Summary-

         This research plan intends to spend three years, through design simulation, process optimization, and measurement analysis, to develop a piezoelectric micro ultrasonic sensor process platform with high output power in the air, and use its components to achieve ultra Sonic tactile and gesture recognition. In the first year of this project, we will purchase wafers with cavities of various sizes and shapes that are manufactured using silicon-on-insulator technology to reduce the time delay and manpower caused by process errors in self-made cavities. Cost, from the beginning, a small amount of trial production will be carried out for the basic structure and advanced test structure, so as to learn from the experience and integrate various process equipment. In order to improve the efficiency of designing new components, this project will use finite element method and numerical analysis method to build component models, and use laser Doppler, nanoindentation, optical profile, impedance analysis and other technologies to measure the components. Dynamic model, static model, mechanical characteristics and electrical impedance. At the same time, this research will also analyze the existing ultrasonic tactile and gesture recognition systems made with traditional piezoelectric ultrasonic sensors, understand its acoustic characteristics and working mechanism, and formulate the specifications of the components to be developed in this research. In the second year of this project, more test structures will be produced and the process will be optimized, with particular emphasis on the polarization of piezoelectric materials to obtain the maximum output power. Various analysis models will also be further optimized and compared with different software. In the final year, various process parameters will be sorted out and standardized processes will be produced, and the developed components will be integrated with the multi-channel research ultrasonic imaging system to demonstrate its applicability in ultrasonic tactile and ultrasonic gesture recognition.

Summary-

         The project "Development and optimization of prospective lead-free miniature piezoelectric ultrasonic sensors (PMUT): from materials to components" Apply and create an environmentally friendly system. In this project, we will discuss the preparation and synthesis of lead-free materials, the application and optimization of materials, and the preparation of thin films on flexible substrates. While establishing the preliminary version of the lead-free miniature piezoelectric ultrasonic sensor (PMUT), we will also verify at the same time The performance of the component. In order to achieve the project goals, the team is composed of teams from three different countries, and uses the expertise of each team to communicate with each other and integrate the best results. The team in Taiwan is mainly responsible for the application of piezoelectric materials, the preparation of the first version of miniature piezoelectric sensors, and the performance verification of analog components; the team in Latvia is mainly responsible for the preparation of lead-free materials. They have rich experience in the past. They can prepare lead-free piezoelectric materials with good characteristics; the team in Lithuania currently has mature piezoelectric material analysis technology and is an expert in the field of wide band gap dielectric materials and acoustic analysis. It is believed that under the cooperation of multiple teams, the development of this lead-free material component will be greatly improved.

         The goal of this project is to create a lead-free material miniature piezoelectric ultrasonic sensor that can be widely used in biomedicine and other fields. We use a variety of preparation methods, such as: pulsed lightning deposition system, scraper walking system, to create more applicability of components, we believe that we can complete a piezoelectric ultrasonic component platform with excellent characteristics.

Still in planning

Summary-

         The purpose of this research project is to explore how to use a commercially available single-channel ultrasonic sensor with a self-designed sonic scanning mirror to make a wearable ultrasonic imaging system for animal experiments. It is expected to use SolidWorks assistance to design the structure by itself to construct the required Acoustic Scanning Mirror body, and then implement it with 3D printing on Phrozen machine, install motors, magnets and coils to make a precise and stable system body, and use Labview and Arduino The collocation controls the motor and captures the data from the miniature ultrasonic sensor, collects and aggregates the obtained data to create a two-dimensional ultrasonic image. It is expected that a miniaturized ultrasound imaging device will be produced and applied in animal experiments. In addition to saving the time of personnel operation on ultrasound detection, as well as reducing the difficulty and cost of research data acquisition, it also allows researchers to observe experimental animals in real time. The latest ultrasound images.

Summary-

         The purpose of this research project is to construct an ultrasonic holographic projection system that can be easily manipulated and programmed, using an ultrasonic sensor array to generate multiple focal points in the air, creating several low-pressure areas to suspend polystyrene particles in the air, and Use a preset program to control the floating position and moving direction of each particle, and arrange them into a desired pattern in a three-dimensional space, thereby establishing an ultrasonic three-dimensional holographic projection image system. In order to focus on the feasibility study, Ultrahaptics’ UHEV1 ultrasonic sensor module will be used as the hardware architecture and Unity, which has been commonly used in various virtual reality games in recent years, as the development engine. The sound wave holographic projection platform provides a new choice of holographic imaging system.

Summary-

        In the blast furnace ironmaking production process, the operator currently needs to sense the temperature distribution of the gas in the furnace through the thermocouple on the temperature measuring rod. However, the existing temperature measuring rod needs to be actually extended into the furnace, so it is easy to cause the material flow trajectory When distributing materials, the material surface depression can often be observed under the temperature measuring rod. This phenomenon will cause uneven distribution of airflow in the furnace, which will affect the efficiency of the steelmaking furnace operation, and the charge will also have an opportunity to collide when feeding. The thermocouple is damaged when the temperature measuring rod is reached, and the length of the temperature measuring rod is as long as five meters. Therefore, it takes a lot of manpower and time to replace it. From the above problems, it can be seen that the temperature measuring rod is used to measure the gas temperature in the furnace. Sensing is not in line with the current development direction of smart high-temperature furnaces, so it is necessary to find better alternatives.

         Based on the above-mentioned difficulties encountered by the existing two-dimensional temperature measurement methods, this project is expected to continue the sound wave temperature measurement method planned in the previous period. Based on the relationship between the sound wave transmission speed and the change of the medium temperature image, the gas temperature in the furnace is calculated and matched with the sound wave The sensor array is used to form the sound wave speed of dozens of sound wave paths in the furnace, and calculate the average sound speed of the gas on each path. Finally, through the development of reconstruction algorithm, the sound speed of dozens of paths is converted into two-dimensional The temperature distribution of the plane gas field is provided for the on-site personnel to make reference for the temperature distribution control in the furnace. In the previous case, the two-dimensional distribution of gas temperature can be successfully reconstructed under the ideal model. However, there are still some technical difficulties in the actual sound wave temperature measurement experiment that need to be overcome by the optimization of software and hardware. Therefore, this phase of the plan is The work mainly includes:

1)開發不受高溫區域大小與形狀影響之聲速計算演算法。此演算法將能根據系統得到之脈衝回聲訊號正確計算出該路徑之聲速，不被聲速傳遞時因高溫氣場瞬息變化產生之不穩定干涉型態所影響。

2)優化聲波訊號擷取電路，增加訊雜比，使聲波計算演算法有更穩定可靠的資料來源。

3)架設非聲波式二維氣體溫度分布感測系統，以驗證本聲波二維氣體溫度分布感測系統之可靠度。

4)開發能將現有37個熱電偶感產生之數據轉換成二維平面溫度分布之演算法，並根據現場需求設計人機介面程式。

5)進行高爐現有之汽笛脈衝回波系統與上期計畫開發之超音波二維溫度分布感測系統之硬體整合

6)蒐集並分析高爐運作時爐內常見之雜訊，並設計可成功於高爐內環境下接收與發射之聲波傳感器。

7)優化於上期計畫中開發之可在實驗室操作之小型二維超音波陣列平面溫度感測系統，驗證第一項開發之聲速計算演算法對超音波二維平面氣場分布演算法的準確度之提升。

8)優化於上期計畫中開發之可在實驗室操作之小型二維超音波陣列平面溫度感測系統，使其能及時顯示二維氣場溫度之變化，並與第三項開發之二維氣場溫度分布驗證系統做比較以評估其可靠度。

Acting as Executive Secretary

Summary-

         In the process of blast furnace ironmaking, the operator uses the thermocouple on the temperature measuring rod to sense the two-dimensional gas temperature distribution in the furnace. However, because the temperature measuring rod needs to be actually extended into the furnace, the material flow trajectory will be destroyed. When distributing materials, it is often observed that the material surface is recessed under the temperature measuring rod, which makes the distribution of the air flow in the furnace uneven, which affects the efficiency of the steelmaking furnace operation, and the charge will hit the temperature measuring rod when the material is fed, which often results in thermoelectricity. Occasionally, the length of the temperature measuring rod is about five meters, so it takes a lot of manpower and time to replace it.

         According to the above phenomenon, there are difficulties encountered by the two-dimensional temperature measurement method. This project is expected to use the ultrasonic temperature measurement method to calculate the temperature of the gas in the furnace based on the relationship between the transmission speed of the sound wave and the change of the temperature of the medium, and cooperate with the ultrasonic array The use of this method is to construct dozens of ultrasonic paths in the furnace and calculate the average temperature of the gas on each path by calculating the sound wave velocity of each path. Finally, through the development of reconstruction algorithm, the temperature of the dozens of paths is calculated. It is converted into a two-dimensional plane gas field temperature distribution, which provides a reference for the on-site personnel to control the temperature distribution in the furnace. The work of this case mainly includes:

1)開發超音波二維平面氣場分布重建演算法。此演算法將根據數十條超音波傳遞路徑量測的結果，運算出各路徑之聲波平均速度，  從而得知各路徑之平均溫度，最後再根據各路徑之溫度計算出各區域之氣體溫度，並繪製平面氣場溫度分布圖形。

2)結合商業化矩陣演算軟體MATLAB以及其聲波傳遞模擬模組K-wave，建立一套二維超音波陣列溫測模擬程式，藉由可編程之超音波探頭數目、超音波探頭位置分配、爐內氣體溫度之分佈等，驗證上述開發之超音波二維平面氣場分佈驗算法的可靠度，其中氣體溫度之分佈除可自行設計外，亦可匯入現有測溫桿之資訊做演算法準確度比對。

3)評估可成功於爐內環境下接收與發射超音波訊號之規格，並協助超音波探頭之安裝與訊號測試。

4)建立一可在實驗室操作之小型二維超音波陣列平面溫度感測系統，驗證第一項開發之超音波二維平面氣場分佈演算法的準確度，並同時開發第三項工作完成後，各超音波探頭的控制電路、硬體介面、以及控制程序之撰寫，並可用以評估於高爐安裝之超音波平面溫度感測系統得穩定度與偵測速度。

Summary-

         This course "Application of Industrial Internet of Things Technology in Smart Aquaculture" is one of the few textbooks that applies cross-field technologies such as the Internet of Things, hardware circuit design, deep learning, and ultrasonic systems to aquaculture, and both course modules include Three units that can be independently promoted can cover the needs of students in different fields and further enhance the opportunities for the promotion of teaching materials. The promotion highlights and expected results of the two modules are as follows:

• The establishment and integration of smart aquaculture Internet of Things (Professor Kwong Xianrong, Department of Information Engineering, Sun Yat-sen University)

This course module contains cross-domain knowledge and technology. It not only introduces the Internet of Things, hardware circuit design, deep learning and other technologies, but also builds a smart aquaculture that can help solve traditional aquaculture problems through a series of practical units. The Internet of Things system is suitable for those who want to learn the basics of the Internet of Things, hardware circuit design, and those who want to learn the basics of deep learning. Since the textbook covers a wide range of fields and can be divided into three parts for independent promotion, good promotion effects can be achieved.

• Ultrasonic system and design practice applied to smart aquaculture (Professor Huang Zhixian, Department of Electrical Engineering, Chenggong University)

The highlight of the promotion of this course module is the few domestic laboratories for the development of ultrasonic systems for aquaculture. In addition to providing an environment for the actual operation of the unit, the laboratory staff will provide developers for the development of new aquaculture ultrasonic technology Consultation and equipment use guidance, laboratory equipment will also be provided for developers to pay (only for equipment that requires regular maintenance and calibration) or free use. It is expected that 10-15 engineers with the ability to design and build ultrasonic systems will be trained every year, and 2 to 3 innovative aquaculture ultrasonic technologies will be developed.

Summary-

         電阻抗成像(Electrical Impedance Tomography)系統優點為可非侵入式的快速產生出人體內部影像，特別常應用於在活動時阻抗變化明顯的器官如肺臟等，但目前EIT系統能為普遍使用，主要原因為成像準確度與X-Ray、EBT等比較起來尚有些差距。不同EIT系統之硬體部分往往大同小異，然而在訊號的處理與成像上則各有不同，特別是演算法的部分，故本計畫將針對演算法部分進行比較與改善。

         本計畫案之重點為著重在比較各種現有之EIT成像演算法，根據高頻呼吸器對肺部特定區域造成之阻抗變化，針對此區域最佳化成像速度與影像準確度以滿足未來臨床實驗之需求，並藉由分析不同演算法之優劣，找出下一階段自行開發EIT成像演算法之改善方向。本案工作主要包含：

1)參考德國學者Bartłomiej Grychtol 於2014年發表的” Functional Validation and Comparison Framework for EIT Lung Imaging” 中的EIT演算法分類，使用EIDORS 與 Netgen 將量測所得資料在不同的肺部模型與逆問題求解方法下做成像並與實驗之結果做比較。

2)與負責訊號實測部分的負責團隊互相配合，討論需要的量測數據，如不同大小的肺部模擬實驗，二維的電極資料等，目的在於先始可控制且已知變化量的擬真實驗進行成像演算法的準確比對，並針對準確度與成像速度最佳化。

3)將深入探討 : 肺部呼吸時造成的胸腔大小改變導致電極位置與角度產生之變化、高頻呼吸器實驗著重之區域進行肺部模型建立的特化、高頻呼吸器實驗造成之導電率變化模式行逆問題求解方式篩選、以及電極排列與三維肺部影像成像的相關性。

Background and motivation-

       In the past decade, the rapid development of the Internet of Things (IoT) and artificial intelligence (AI) has increased the demand for multifunctional sensors and actuators. Due to the maturity of microelectromechanical systems (MEMS) technology, the most advanced MEMS sensors dominate the sensor/actuator market. However, as the world’s leading semiconductor manufacturing country, Taiwan’s MEMS suppliers have a significantly lower market share. Therefore, Taiwan should copy the successful experience of the semiconductor and IC design industries and seek breakthroughs. Point to inspire the further development of the MEMS industry.

At present, the most common application of ultrasound sensors is ultrasound imaging systems. In recent years, it has played an important role in diagnostic imaging such as non-destructive imaging, deep tissue imaging and real-time imaging. However, traditional ultrasound sensors made of piezoelectric materials The size limits its feasibility in IoT applications.

       Miniature ultrasonic sensor technology, especially piezoelectric miniature ultrasonic sensor (pMUT), because pMUT drives a suspended thin film to vibrate through a piezoelectric material layer, it has a small size, a large acoustic impedance matching with air/water, and a low driving voltage. It has attracted wide attention due to its advantages such as small size and low drive voltage. It is easy to be embedded in hardware with a small space. The low acoustic impedance mismatch can achieve higher frequency applications (> 250KHz), which in turn brings higher resolution. Spend. However, there are still some challenges to promote the technology to compete with the existing piezoelectric transducer technology and apply the viewpoints of the Internet of Things industry, such as cost-effective mass production and reliable quality. The main purpose of this proposal is to cooperate with TSRI experts and train NCKU students to jointly develop a reliable pMUT platform that can meet the requirements of IoT applications.

研究方法-

第一年

In the first year, the prototype of the piezoelectric micro ultrasonic sensor (pMUT) platform will be constructed based on the structure of PZT, AlN or PVDF stacked on a thin film formed of silicon dioxide. The development of the pMUT platform will include three main aspects:

(1) The polarization process of piezoelectric materials

Regarding the polarization process, the performance of pMUT is mainly affected by the parameters of piezoelectric materials. Compared with other materials, PZT has a larger piezoelectric coefficient. Therefore, PZT should be the most efficient piezoelectric material in this research and can produce The maximum output sound pressure, despite this, the polarization results on the prefabricated film/post-fabricated film on the wafer are still not clear for different piezoelectric materials. In this project, a comprehensive study of the different polarization processes of each piezoelectric material will be carried out to further understand its mechanical principles.

(2) Back etching process for preparing thin film

The acoustic characteristics of pMUT mainly depend on the mechanical characteristics of the element structure layer. When the array elements increase to thousands, the back etching process used to prepare the film plays an important role in the consistency of the response frequency and the performance of the ultrasonic system. In this proposal, efforts will be made to study the process technology and mask design that can achieve high aspect ratio etching to ensure that the back etching results are consistent with the expected design.

(3) Simulation model and finite element analysis of pMUT

In order to predict the preparation results and optimize the process parameters for the required specifications of the components, a finite element analysis model of the same material layer as the designed pMUT will be established. This model has axial symmetry and considers the conversion relationship between electrical and mechanical. The mechanical properties of electrical materials will be measured by a nanoindentation mechanical property analyzer, and the piezoelectric charge constant (d31) can be measured by a laser interferometer.

the second year

Based on the pMUT platform prototype built in the previous year, in the second year, it will focus on optimizing the process flow and formulating design standards. At the same time, it will also improve the pMUT structure to improve performance:

(1) Reliability of back etching process and polarization

In order to apply pMUT to different Internet of Things, a reliable platform is needed to provide researchers to modify their component designs.

In terms of back etching, in addition to cavities, the technology of back etching will be further studied to prepare holes with different requirements such as trenches and through holes; in terms of polarization, the polarization process under different conditions will be evaluated, especially for integration The restricted process, for example, the high temperature of RTA and the thermal polarization used in PZT will limit its process integration with CMOS and material selection.

(2) Establish design standards for pMUT

Based on the measurement results and the FEM model, a comprehensive analysis will be carried out to define the design standards of the pMUT platform. Through the design process, research can quickly manufacture pMUT devices specifically for specific IoT requirements.

(3) Structure to optimize power efficiency

In recent years, studies have pointed out that the damping coefficient can be adjusted by changing the structure around the vibrating device. In order to achieve the acoustic characteristics required by specific IoT applications, the basis and system research of optimizing the quality factor is a necessary topic.

The third year

Last year, different pMUTs were designed for different IoT applications. This year, their characteristics will be measured and optimized through design and process. A display system will be established to improve the future benefits of the industry. Detailed process flow and design standards will be proposed, and the future will be provided. A guide for researchers on how to use this pMUT platform.

(1) Prepare pMUT components for different IoT applications

According to the sensors or actuators required by specific IoT applications, the pMUT platform developed by this project will be able to produce a variety of pMUTs with good performance, and publish classification reports of these components to distinguish the performance results.

(2) Develop display system

Use the above-mentioned pMUT components to establish a reliable proof-of-concept system for different IoT applications.

(3) Deliver the design and process agreement pMUT platform separately

Organize and file the pMUT platform including design standards and FEM models

Background and motivation-

        最近，物聯網(IoT)的蓬勃發展提升了感測、驅動與成像的需求，由於微機電系統(MEMS)技術的長期發展，MEMS傳感器主導了感測器/致動器的市場，然而，作為全球領先的半導體製造國，台灣對MEMS技術的研究相對較少，因此台灣應複製半導體和IC設計產業的成功經驗，找尋突破點以啟發MEMS產業的進一步發展。

        超音波傳感器已廣泛應用於非破壞醒檢測、指紋辨識、細胞處理和醫學成像中，傳統的超音波傳感器由d33厚度模式塊狀壓電陶瓷製成，其共振頻率直接取決於壓電層厚度，限制了設計的靈活性以及其應用範圍；相反的，壓電式微型超音波傳感器(pMUT)提供了一種替代的方法來製造具有高頻和靈活尺寸的超音波傳感器，此外，pMUT具有功耗低、頻寬高、成本低以及易於製造大型陣列的優點，由於這些優勢，已有研究團隊將PMUT應用於測距、指紋辨識和光聲成像中。

        目前已有多種壓電材料用於製作pMUT，包括聚偏二氟乙烯(PVDF)，鋯鈦酸鉛(PZT)和氮化鋁(AlN)等，由於PZT具有較高的壓電常數，使用PZT陶瓷將顯著提升壓電式MEMS元件的表現，然而，將PZT薄化至50 µm以下與製程整合極具挑戰性，因此製作pMUT中很少使用PZT陶瓷，同時在薄化PZT的過程中，發現了包括孔隙率，碎裂和晶圓級均勻性在內的缺陷，但是，厚度小於10 µm的PZT陶瓷在製造具有特定頻率的微型元件中仍是必須的，尤其是在製備pMUT陣列上更為重要。

        此提案的目標為開發一種新的濺鍍PZT方法，製備出厚度2µm以下且具有極佳壓電特性的PZT，並可應用於pMUT上，與當前的常規塊狀陶瓷PZT或溶膠-凝膠PZT相比，使用濺鍍沉積可以提高可靠性和製程整合性。目前，TSRI的研究員正在與NCKU合作，透過NCKU進行材料分析和元件模型模擬來優化PZT薄膜的特性，本計畫能夠建立一個可靠的pMUT平台，以滿足IoT應用的需求。

研究方法-

        在未來一年中，將建立一個基於濺鍍PZT的壓電式微型超音波傳感器(pMUT)平台，在已有圖案的SOI晶片上建立結構層，pMUT平台的開發包含四個主要方面:

(1) 壓電材料特性分析：

關於壓電特性的表現，pMUT的性能主要受壓電材料的參數所影響，與其他製備方法相比，濺鍍PZT最容易與其他製程整合，因此被認為是製造pMUT最有效的方法，儘管如此，沉積層仍然需要通過不同測量方法進行優化，包括XRD，SEM和PE曲線。在此計畫中，將對濺鍍PZT薄膜進行包括不同方面的全面研究，以從根本上進一步理解機械原理。

(2) Back etching process for preparing thin film

The acoustic characteristics of pMUT mainly depend on the mechanical characteristics of the element structure layer. When the array elements increase to thousands, the back etching process used to prepare the film plays an important role in the consistency of the response frequency and the performance of the ultrasonic system. In this proposal, efforts will be made to study the process technology and mask design that can achieve high aspect ratio etching to ensure that the back etching results are consistent with the expected design.

(3) Simulation model and finite element analysis of pMUT

In order to predict the preparation results and optimize the process parameters for the required specifications of the components, a finite element analysis model of the same material layer as the designed pMUT will be established. This model has axial symmetry and considers the conversion relationship between electrical and mechanical. The mechanical properties of electrical materials will be measured by a nanoindentation mechanical property analyzer, and the piezoelectric charge constant (d31) can be measured by a laser interferometer.

(4) 利用PZT濺鍍技術優化pMUT特性

pMUT的性能與其結構層、半徑和壓電係數的組合具有高度相關，透過將模擬結果和測量及果相互匹配，建立設計流程，以快速評估在不同共振頻率或輸出聲壓下所需的參數。在此計畫中，第一步將嘗試不同的厚度及尺寸來優化PZT濺鍍層的性能，對於具有不同厚度和半徑的PZT層，將評估其機電耦合係數(k31)，接著，將使用模擬模型提出針對特定物聯網應用所需之共振頻率的初步設計，製備出的pMUT將利用LDV和標準麥克風進行特性分析，以修改此平台的聲機械耦合係數，最後，將對所有TSRI用戶建立並開放使用濺鍍PZT及其設計標準的pMUT平台。