Temperature measurement-based production logistics planning system for research and development of thermosensitive chips

Abstract

The invention provides a temperature measurement-based production logistics planning system for research and development of thermosensitive chips. The temperature control safety and the logistics scheduling intelligent level of the thermosensitive chips in the multi-node circulation process are improved; the method comprises the following steps: firstly, deploying temperature sensors in a material storage area, a transportation path node and an equipment interface, collecting environment temperature-time sequence data, constructing a temperature fluctuation feature vector, and calculating a thermal stress equivalent value based on a material response function; thirdly, a constraint rule is established according to parameters such as thermal exposure duration and path transportation time efficiency of each node, and multi-objective optimization is fused to generate an optimal logistics path meeting thermal threshold limitation; and finally, when the path thermal risk is close to the material tolerance, the system intelligently judges and executes local temperature control or path reconstruction, and whole-process closed-loop adjustment is realized. Dynamic temperature control and path cooperative control in the thermosensitive chip transportation process are achieved, and the thermal safety, path intelligence and production yield of thermosensitive chip transportation are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of heat-sensitive chip logistics planning, in particular to a production logistics planning system for heat-sensitive chip research and development based on temperature measurement. BACKGROUND

[0002] Heat-sensitive chips are a class of semiconductor devices that are extremely sensitive to temperature changes. The material properties of heat-sensitive chips will change significantly due to the fluctuation amplitude, abnormal duration, and fluctuation frequency of the environmental temperature. During the research and development and production processes, the chips need to go through multiple node logistics links, including material storage, production line transfer, and equipment docking. Any temperature control deviation can easily cause material thermal stress accumulation, leading to performance degradation or even failure. The production logistics planning system monitors the temperature data in the transportation path of heat-sensitive chips in real time to ensure that the heat-sensitive chips are in a safe transportation environment throughout the logistics cycle.

[0003] However, the current heat-sensitive chip production logistics planning system still faces some challenges in actual operation. First, the conventional heat-sensitive chip production logistics planning system mostly uses a static temperature threshold judgment method, ignoring the influence of temperature change trend and time dimension, making it difficult to comprehensively evaluate the temperature control risk. Second, the conventional method only targets the shortest path or the fastest time efficiency, ignoring the cumulative effect of thermal stress risk on the entire path, resulting in frequent selection of high-temperature environment paths and increasing the failure rate. Finally, the traditional system cannot adjust the temperature control strategy or change the path in time according to the predicted value of thermal stress, and the response capability to high-temperature emergencies is insufficient. SUMMARY

[0004] To solve the technical problems mentioned in the background art, the purpose of the present application is to provide a production logistics planning system for heat-sensitive chip research and development based on temperature measurement.

[0005] To achieve the above-mentioned purpose, the present application adopts the following technical solutions: A production logistics planning system for heat-sensitive chip research and development based on temperature measurement, comprising: M1, a temperature field perception module, deploying temperature sensors at material storage areas, transmission path nodes, and production equipment interface positions to collect real-time temperature-time sequence raw data of the environment in which the heat-sensitive chips are located; M2, a temperature-sensitive constraint modeling module, composed of a waveform analysis unit and an operation unit. The waveform analysis unit is used to extract mathematical features from the temperature-time sequence raw data, output temperature fluctuation three elements, and obtain a temperature fluctuation feature vector based on the temperature fluctuation three elements. The operation unit converts the temperature fluctuation feature vector into an equivalent value of cumulative thermal stress through a material response function, and generates node constraint rules through the equivalent value of cumulative thermal stress; M3, a dynamic path planning module, according to the node constraint rule, real-time tracking the heat exposure history length and temperature change frequency data of the material flow through the path node, and taking the path transportation time cost and the cumulative heat stress equivalent value as the synchronous optimization target, generating the optimal logistics path meeting the node constraint rule through the multi-objective function; M4, an intelligent response execution module, when detecting that the cumulative heat stress equivalent value of the optimal logistics path is close to the material threshold value of the heat-sensitive chip, executing a response mechanism according to the availability of the current path segment temperature control equipment.

[0006] Further, the deployed temperature sensors are numbered as , wherein is the total number of deployed temperature sensors, and for the temperature sensor numbered , the temperature-time sequence original data is represented as: , wherein is the temperature-time sequence collected by the th sensor; is the timestamp of the th sampling, in seconds (s); is the temperature sampling value of the th temperature sensor at time , in degrees Celsius; is the total number of sampling points of the sensor .

[0007] Further, the three elements of temperature fluctuation include temperature change amplitude, temperature abnormality duration, and temperature fluctuation frequency, the time window length is , the window start time is , and the temperature change amplitude, temperature abnormality duration, and temperature fluctuation frequency are extracted in the time window interval . The extraction formula of the temperature change amplitude is: , wherein represents the maximum numerical difference of temperature change in the time window interval , as a quantitative indicator of temperature fluctuation amplitude; and represent the maximum value and the minimum value in the time window interval , respectively; The extraction process of the temperature abnormality duration is the calculation of the temperature mean value in the time window interval , the determination of the temperature abnormality state, the definition of the temperature abnormality interval, and the definition of the temperature abnormality duration; The time window interval The calculation formula of the internal temperature mean value is: Wherein, is the average value of the temperature sampling points in the window; is the number of sampling data points in the current window; The determination condition of the temperature abnormal state is: Wherein, is the deviation value of the sampling point temperature and the average value of the temperature sampling points in the window; is the temperature difference abnormality determination threshold; Extract all time intervals that continuously satisfy the temperature abnormal state determination condition, and define the time interval with the longest duration as the abnormal interval , and Wherein, and are the start time and end time of the abnormal interval respectively; Based on the abnormal interval Calculate the time length of the interval, and define the temperature abnormality duration as the temperature abnormality duration , unit: second, and .

[0008] Further, the extraction process of the temperature fluctuation frequency is the definition of the continuous temperature data sequence, the calculation of the number of times of passing through the homogenization in the temperature sequence, and the calculation of the temperature fluctuation frequency; The continuous temperature data set collected by the sensor in the time window is denoted as , and Wherein, is the total number of temperature sampling points of node in the first sliding window; Project the continuous temperature data sequence on the time axis as the temperature sequence , so as to calculate the number of times of passing through the homogenization in the temperature sequence, and the calculation formula is: Wherein ， represents a pass; is an indicator function, which is 1 when the condition is true, and 0 when the condition is false; is the number of times of passing through the homogenization mean value of the temperature sequence in the current window; Calculate the temperature fluctuation frequency based on the number of times of passing through the homogenization in the temperature sequence, and the calculation formula is: in, is the frequency of temperature fluctuation per unit time; is the time length of the sliding window; The temperature fluctuation feature vector is obtained based on the temperature change amplitude, temperature anomaly duration and temperature fluctuation frequency. ,and .

[0009] Furthermore, the temperature fluctuation characteristic vector Enter the material response function, which is calculated as: in, is the thermal stress equivalent value; is the benchmark adjustment coefficient; is the response index of the temperature fluctuation amplitude; is the response index of the duration of temperature anomaly; is the response index of temperature fluctuation frequency; Calculating a cumulative thermal stress equivalent value based on the thermal stress equivalent value, wherein the calculation process of the cumulative thermal stress equivalent value includes calculating the number of effective sliding windows and calculating the cumulative thermal stress equivalent value; The number of effective sliding windows The calculation formula is: in, is the cumulative time period of thermal stress; is the time length of the sliding window; Based on the effective sliding window number Calculate the cumulative thermal stress equivalent value , the calculation formula is: in, For the Thermal stress equivalent value of each window; Based on the cumulative thermal stress equivalent value Generate the node constraint rules ,and ,in, is the maximum thermal stress threshold allowed by the thermal-sensitive chip material.

[0010] Furthermore, the thermal exposure history duration is the cumulative exposure time of the thermal-sensitive chip during the process of passing through all nodes of the path, and the tracking process of the thermal exposure history duration is the calculation of the effective thermal exposure time and the calculation of the thermal exposure history duration; For each node in the path , the temperature sequence is collected by the deployed temperature sensor , then the effective heat exposure time of the node is: ， in, For the The effective heat exposure time of each node, i.e., the transmission or residence time, in seconds (s); For nodes The total number of temperature sampling points; is an indicator variable; is the temperature sampling time interval; For the The temperature value of each sampling point; is the safety temperature threshold; Based on the effective heat exposure time The heat exposure history duration is calculated using the following formula: in, is the total duration of heat exposure history, in seconds; is the number of nodes traversed in the path.

[0011] Furthermore, the temperature change frequency data is the cumulative number of temperature fluctuations experienced by the thermal chip in the entire path, and the calculation formula is: in, is the total frequency of temperature fluctuations along the entire path, in times; For the The number of times the temperature sequence within a node crosses the mean temperature is used to indicate the severity of the temperature change; The calculation process of the path transport time cost includes calculating the path segment transport time and accumulating the total path segment transport time; The transport time of the path segment is composed of three parts: path length, transport speed and additional delay, and the calculation formula is: in, is the route segment transportation time; For path segments Length, in meters; is the average transport speed on the path segment, in meters per second; The system scheduling delay time is in seconds; Based on the route segment transportation time The total transport time of the path segment is obtained by adding up , the calculation formula is: based on the total length of the heat exposure history , the total sum of the full path temperature fluctuation frequency and the path segment transportation time get the optimal logistics path ， The expression is: Find wherein, is the optimal logistics path; is the path variable to be optimized, which represents the set of nodes and path segments passed by the heat-sensitive chip from the starting point to the ending point; is a multi-objective function; from all paths , find the path that minimizes the objective function.

[0012] Further, based on the cumulative thermal stress equivalent value of the optimal logistics path the thermal stress risk coefficient ratio of the optimal logistics path is obtained: wherein, is the thermal stress risk coefficient ratio, and ; is the maximum thermal stress threshold allowed by the heat-sensitive chip material; compare the thermal stress risk coefficient ratio with the warning threshold to determine whether the response mechanism needs to be triggered; when at this time, it means that the cumulative thermal stress equivalent value is in the safe interval and does not need to be responded; when at this time, it means that the cumulative thermal stress equivalent value is close to the material threshold, and the response mechanism needs to be triggered.

[0013] Further, the response mechanism is divided into the current path segment with temperature control capability and the current path segment without temperature control capability; When the current path segment has temperature control capability, the system starts local temperature control equipment to reduce the current temperature, and the thermal stress prediction value calculation process after temperature control includes the calculation of the predicted reduced thermal stress equivalent value and the generation of the thermal stress prediction value after temperature control adjustment. The calculation formula of the predicted reduced thermal stress equivalent value is: wherein, is the predicted reduced thermal stress equivalent value; is the temperature control efficiency coefficient; is a current temperature of the path segment; is a target temperature control temperature; based on the predicted reduction of the thermal stress equivalent value calculating the thermal stress prediction value after temperature control adjustment, the calculation formula is: wherein, is the thermal stress prediction value after temperature control adjustment; is a safety proportion coefficient; When the current path segment does not have temperature control capability or the thermal stress prediction value after temperature control adjustment is still over standard, the system will call the dynamic path planning module M3, eliminate the current path segment in the total path segment and regenerate the optimal logistics path of the cumulative thermal stress equivalent value constraint.

[0014] Compared with the prior art, the present application has the following advantages: 1、The present application overcomes the drawbacks of the prior art, such as relying on static temperature threshold, ignoring time dimension and dynamic trend, by proposing a feature extraction method of temperature fluctuation three elements, i.e. temperature change amplitude, temperature abnormality duration and temperature fluctuation frequency, and combining material response function to obtain cumulative thermal stress equivalent value, which not only improves the accuracy of thermal stress evaluation, but also makes the evaluation result have physical meaning at material level, and can provide more reliable data support for chip material design and process matching, and is suitable for new type of semiconductor materials which are extremely sensitive to thermal conditions.

[0015] 2、The present application jointly includes path transportation time cost, total thermal exposure history duration, full path temperature fluctuation frequency and cumulative thermal stress equivalent value into the dynamic logistics path planning optimization model, calculates the optimal path through multi-objective function, realizes synchronous optimization of transportation efficiency and thermal stress safety, compared with the traditional path planning method which only takes the shortest time as the target, this mechanism can dynamically avoid high temperature nodes and adverse environment segments, prevent the performance degradation of chips caused by excessive thermal exposure during transportation, reduce the failure rate of thermal sensitive chips caused by transportation temperature control failure, and improve the overall yield and batch stability of thermal sensitive chips.

[0016] 3、The application can determine whether the current path is close to the thermal sensitive chip material tolerance in real time by continuously monitoring the accumulated thermal stress equivalent value in the chip running path through the intelligent response execution module, and introducing the risk coefficient ratio and early warning threshold mechanism, and then dynamically triggering the local temperature control or path reconstruction strategy, if the current path segment has temperature control capability, the system can automatically adjust the temperature control parameters for thermal stress adjustment, if it does not have temperature control capability, the system restarts the path planning module, eliminates the high-risk path segment and generates a new path, realizes the complete closed-loop thermal risk control process from "perception-modeling-warning-response", greatly improves the response ability of the system to sudden thermal environment changes, and has good practicality and engineering expansibility. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0018] Figure 1 The system workflow schematic diagram of the present application; Figure 2 The node constraint rule generation process schematic diagram of the present application; Figure 3 The optimal logistics path generation process schematic diagram of the present application. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical scheme and advantages of the embodiments of the present application more clear, the technical solutions in the embodiments of the present application will be described clearly and completely in the following with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor belong to the scope of protection of the present application.

[0020] In order to achieve the above purpose, the present application is realized by the following technical scheme, and the present application provides a production logistics planning system for thermal sensitive chip research and development based on temperature measurement, as shown in the figure, the system comprises: Figures 1-3 M1, temperature field perception module, temperature sensors are arranged at material storage area, transmission path node and production equipment interface position, to collect temperature-time sequence original data of the environment where the thermal sensitive chip is located in real time.

[0021] The positions of the material storage area which need to arrange temperature sensors include raw material warehouse, waiting processing area, finished product buffer area, etc., to monitor the temperature accumulation during the material standing;​ The positions where the temperature sensors need to be deployed in the transport path nodes include automatic handling paths, rail junctions, conveyor belt connection areas, elevators, chute transfer sections, etc., to monitor the dynamic temperature exposure of materials during movement; The positions where the temperature sensors need to be deployed in the production equipment interface positions include welding machines, packaging lines, heat treatment equipment, test equipment import and export, etc., to capture the short-term impact of process heat sources on materials; The multiple temperature sensors deployed are numbered as , wherein is the total number of deployed temperature sensors, and for the temperature sensor numbered , the temperature-time sequence raw data is represented as: , wherein is the temperature-time sequence collected by the th sensor; is the timestamp of the th sampling, in seconds (s); is the temperature sampling value of the th temperature sensor at time , in degrees Celsius (℃); is the total number of sampling points of the sensor .

[0022] M2, a temperature-sensitive constraint modeling module, is composed of a waveform analysis unit and an operation unit, the waveform analysis unit is used to perform mathematical feature extraction on the temperature-time sequence raw data, output temperature fluctuation three elements, and obtain a temperature fluctuation feature vector based on the temperature fluctuation three elements; the operation unit converts the temperature fluctuation feature vector into a cumulative thermal stress equivalent value through a material response function, and generates a node constraint rule through the cumulative thermal stress equivalent value.

[0023] The temperature fluctuation three elements include temperature change amplitude, temperature abnormal duration, and temperature fluctuation frequency; In this embodiment, the time window length is defined as , the window start time is , and the temperature change amplitude, temperature abnormal duration, and temperature fluctuation frequency are extracted in the time window interval ; The extraction formula of the temperature change amplitude is: , wherein represents the maximum numerical difference of temperature change in the time window interval , which is used as a quantitative indicator of temperature fluctuation amplitude; and representing the maximum and minimum values respectively in the time window interval ; The extraction process of the temperature anomaly duration is the calculation of the temperature mean value in the time window interval , the determination of the temperature anomaly state, the definition of the temperature anomaly interval, and the definition of the temperature anomaly duration; The calculation formula of the temperature mean value in the time window interval is: wherein, is the average value of the temperature sampling points in the window; is the number of sampling data points in the current window; In this embodiment, the determination condition of the temperature anomaly state is defined as: , wherein, is the deviation value of the sampling point temperature and the average value of the temperature sampling points in the window; is the temperature difference anomaly determination threshold; Extract all time intervals that continuously satisfy the above-mentioned temperature anomaly state determination condition, and define the time interval with the longest duration as the anomaly interval , and wherein, and are the start time and end time of the anomaly interval respectively; Based on the anomaly interval , the time length of the interval is calculated, and is defined as the temperature anomaly duration , with the unit of second (s), and ; The extraction process of the temperature fluctuation frequency is the definition of the continuous temperature data sequence, the calculation of the number of traversing homogeneity in the temperature sequence, and the calculation of the temperature fluctuation frequency; In this embodiment, the continuous temperature data set collected by the sensor in the time window is denoted as , and wherein, is the total number of temperature sampling points of node in the first sliding window; The continuous temperature data sequence is projected as a temperature sequence on the time axis, so as to calculate the number of traversing homogeneity in the temperature sequence, and the calculation formula is: wherein, represents a once crossing; * is a characteristic function, 1 for true condition, 0 for false condition; is the number of times of crossing the mean value in the current window temperature sequence; The temperature fluctuation frequency is calculated based on the number of times of crossing the mean value in the temperature sequence, and the calculation formula is: wherein, is the temperature fluctuation frequency in unit time; is the time length of the sliding window; In the embodiment, the time length of the sliding window is The value of is 60, and the unit is second (s); The temperature fluctuation feature vector is derived based on the three features of the temperature change amplitude, the temperature anomaly duration and the temperature fluctuation frequency, and ; The temperature fluctuation feature vector is input into the material response function to estimate the degree of thermal stress influence of the current temperature condition on the chip material, and the calculation formula is: wherein, is the thermal stress equivalent value; is the reference adjustment coefficient; is the response index of the temperature fluctuation amplitude; is the response index of the temperature anomaly duration; is the response index of the temperature fluctuation frequency; In the embodiment, the value of the reference adjustment coefficient is ; the value of the response index of the temperature fluctuation amplitude is 2.0; the value of the response index of the temperature anomaly duration is 1.5; and the value of the response index of the temperature fluctuation frequency is 1.0; The cumulative thermal stress equivalent value is calculated based on the thermal stress equivalent value to quantify the total thermal exposure duration of the heat-sensitive chip material from the beginning of the path to the current time, and the calculation process of the cumulative thermal stress equivalent value is the calculation of the effective sliding window number and the calculation of the cumulative thermal stress equivalent value; The calculation formula of the effective sliding window number is: wherein, is the thermal stress accumulation time period; is the time length of the sliding window; In this embodiment, the thermal stress accumulation time period The value is 1800, in seconds (s); the sliding window time length The value is 60, in seconds (s); Based on the effective sliding window number Calculate the cumulative thermal stress equivalent value , the calculation formula is: in, For the Thermal stress equivalent value of each window; Based on the cumulative thermal stress equivalent value Generate the node constraint rules ,and ,in, is the maximum thermal stress threshold allowed by the thermal chip material; In this embodiment, the maximum thermal stress threshold allowed by the thermal chip material is The value of is 2.5.

[0024] M3, dynamic path planning module, tracks the historical heat exposure duration and temperature change frequency data of materials flowing through path nodes in real time according to the node constraint rules, and uses the path transportation time cost and the accumulated thermal stress equivalent value as the synchronous optimization goals, and generates the optimal logistics path that meets the node constraint rules through a multi-objective function.

[0025] The heat exposure history duration and temperature change frequency data are used to reflect the heat exposure risk accumulation status of the heat-sensitive chip along the entire path from the starting point to the target node; The thermal exposure history duration is the cumulative exposure time of the thermal-sensitive chip during the process of passing through all nodes of the path, and the tracking process of the thermal exposure history duration is the calculation of the effective thermal exposure time and the calculation of the thermal exposure history duration; For each node in the path , the temperature sequence is collected by the deployed temperature sensor , then the effective heat exposure time of the node is: ， in, For the The effective heat exposure time of each node, i.e., the transmission or residence time, in seconds (s); For nodes The total number of temperature sampling points; is an indicator variable; is the temperature sampling time interval; For the The temperature value of each sampling point; is the safety temperature threshold; In this embodiment, the safety temperature threshold The value is 25℃; Based on the effective heat exposure time The heat exposure history duration is calculated using the following formula: in, is the total duration of heat exposure history, in seconds (s); is the number of nodes traversed in the path; The temperature change frequency data is the cumulative number of temperature fluctuations experienced by the thermal chip in the entire path, and the calculation formula is: in, is the total frequency of temperature fluctuations along the entire path, in times; For the The number of times the temperature sequence within a node crosses the mean temperature is used to indicate the severity of the temperature change; The calculation process of the path transport time cost includes calculating the path segment transport time and accumulating the total path segment transport time; The transport time of the path segment is composed of three parts: path length, transport speed and additional delay, and the calculation formula is: in, is the route segment transportation time; For path segments Length, in meters; is the average transport speed on the path segment, in meters per second; The system scheduling delay time is in seconds; Based on the route segment transportation time The total transport time of the path segment is obtained by adding up , the calculation formula is: Based on the total duration of the heat exposure history The total frequency of temperature fluctuations along the entire path and the transport time of the route segment Get the optimal logistics path ， The formula is: Find wherein, is the optimal logistics path; is the path variable to be optimized, representing the set of nodes and path segments that the heat-sensitive chip passes through from the starting point to the ending point; is a multi-objective function; represents finding the path that minimizes the objective function from all paths .

[0026] M4, an intelligent response execution module, when detecting that the cumulative thermal stress equivalent value of the optimal logistics path is close to the material threshold value of the heat-sensitive chip, executes a response mechanism according to the current path segment temperature control device availability.

[0027] based on the cumulative thermal stress equivalent value of the optimal logistics path derives the thermal stress risk coefficient ratio of the optimal logistics path: wherein, is the thermal stress risk coefficient ratio, and ; is the maximum thermal stress threshold value allowed by the material of the heat-sensitive chip; compares the thermal stress risk coefficient ratio with a warning threshold value to determine whether a response mechanism needs to be triggered; when , it means that the cumulative thermal stress equivalent value is in a safe interval and does not need to be responded to; when , it means that the cumulative thermal stress equivalent value is close to the material threshold value, and the response mechanism needs to be triggered; In this embodiment, the value of the warning threshold value is 0.8; The response mechanism is divided into the current path segment having temperature control capability and the current path segment not having temperature control capability; When the current path segment has temperature control capability, the system starts a local temperature control device (such as a cooling device, a heat shielding structure, etc.) to reduce the current temperature, thereby slowing down the thermal stress growth trend. The calculation process of the thermal stress prediction value after temperature control includes the calculation of the predicted reduced thermal stress equivalent value and the generation of the thermal stress prediction value after temperature control adjustment. The calculation formula of the predicted reduced thermal stress equivalent value is: wherein, is the predicted reduced thermal stress equivalent value; is the temperature control efficiency coefficient; is the current temperature of the path segment; is the target temperature control temperature; ​In this embodiment, the temperature control efficiency coefficient The value of is 0.02; Based on the projected reduction in thermal stress equivalent values The thermal stress prediction value after the temperature control adjustment is calculated using the following formula: in, is the predicted value of thermal stress after temperature control adjustment; is the safety proportional factor; In this embodiment, the safety ratio coefficient The value of is 0.8; When the current path segment does not have the temperature control capability or the thermal stress prediction value still exceeds the standard after the temperature control adjustment, the system will call the dynamic path planning module M3, eliminate the current path segment from the total path segment and regenerate the optimal logistics path with the cumulative thermal stress equivalent value constraint condition.

[0028] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0029] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A production logistics planning system for the development of thermal chips based on temperature measurement, characterized in that: include: M1, temperature field perception module, deploys temperature sensors in material storage areas, transmission path nodes, and production equipment interface locations to collect real-time temperature-time series raw data of the environment where the thermosensitive chip is located; M2, a temperature-sensitive constraint modeling module, is composed of a waveform analysis unit and a calculation unit. The waveform analysis unit is used to extract mathematical features from the original temperature-time series data, output three temperature fluctuation elements, and derive a temperature fluctuation characteristic vector based on the three temperature fluctuation elements. The calculation unit converts the temperature fluctuation characteristic vector into a cumulative thermal stress equivalent value through a material response function, and generates a node constraint rule based on the cumulative thermal stress equivalent value. M3, a dynamic path planning module, tracks the historical heat exposure duration and temperature change frequency data of materials flowing through path nodes in real time according to the node constraint rules, and uses the path transportation time cost and the accumulated thermal stress equivalent value as the synchronous optimization objectives to generate the optimal logistics path that meets the node constraint rules through a multi-objective function; M4, an intelligent response execution module, when detecting that the accumulated thermal stress equivalent value of the optimal logistics path is close to the material threshold of the thermosensitive chip, executes a response mechanism according to the availability of the temperature control equipment in the current path segment.

2. A production logistics planning system for thermal chip R&D based on temperature measurement according to claim 1, characterized in that: The temperature sensors after deployment are numbered as ,in is the total number of temperature sensors deployed, for The temperature-time series raw data of the temperature sensor is expressed as: in, For the Temperature-time series collected by sensors; No. The timestamp of the subsampling, in seconds (s); For the A temperature sensor at time The temperature sampling value in degrees Celsius; For sensors The total number of sampling points.

3. The production logistics planning system for the development of thermal chips based on temperature measurement according to claim 1 is characterized in that: The three elements of temperature fluctuation include the temperature change amplitude, the duration of temperature anomaly and the frequency of temperature fluctuation. The time window length is , the window start time is , and in the time window Extracting the temperature change amplitude, temperature anomaly duration and temperature fluctuation frequency; The extraction formula of the temperature variation range is: in, Indicates that in the time window The maximum numerical difference of temperature change within the range is used as a quantitative indicator of the temperature fluctuation amplitude; and Indicates that in the time window Take the maximum and minimum values ​​within respectively; The extraction process of the duration of the temperature anomaly is a time window interval Calculation of internal temperature mean, determination of temperature anomaly status, definition of temperature anomaly range, and definition of duration of temperature anomaly; The time window interval The calculation formula for the mean internal temperature is: in, is the average value of the temperature sampling points within the window; is the number of sampled data points in the current window; The determination conditions for the abnormal temperature state are: in, is the deviation between the temperature of the sampling point and the average value of the temperature sampling points in the window; is the temperature difference abnormality judgment threshold; Extract all the time intervals that continuously meet the above-mentioned temperature abnormality judgment conditions, and define the longest duration as the abnormal interval ,and ,in, and are the start time and end time of the abnormal interval respectively; Based on the abnormal interval Calculate the duration of this interval and define it as the duration of the temperature anomaly , in seconds, and .

4. A production logistics planning system for the development of thermal chips based on temperature measurement according to claim 3, characterized in that: The process of extracting the temperature fluctuation frequency is to define a continuous temperature data sequence, calculate the number of homogenization passes in the temperature sequence, and calculate the temperature fluctuation frequency; The sensor In the time window The continuous temperature data set collected in ,and ,in, For nodes In the The total number of temperature sampling points in the sliding window; Project the continuous temperature data sequence onto the time axis as a temperature sequence , used to calculate the number of times the temperature sequence passes through the homogenizer, the calculation formula is: in ， Indicates a crossing; {*} is an indicator function, which is 1 if the condition is true and 0 if the condition is false; is the number of times the temperature series crosses the mean in the current window; The temperature fluctuation frequency is calculated based on the number of times the temperature sequence passes through the homogenizer, and the calculation formula is: in, is the frequency of temperature fluctuation per unit time; is the time length of the sliding window; The temperature fluctuation feature vector is obtained based on the temperature change amplitude, temperature anomaly duration and temperature fluctuation frequency. ,and .

5. A production logistics planning system for the development of thermal chips based on temperature measurement according to claim 4, characterized in that: The temperature fluctuation eigenvector Enter the material response function, which is calculated as: in, is the thermal stress equivalent value; is the benchmark adjustment coefficient; is the response index of the temperature fluctuation amplitude; is the response index of the duration of temperature anomaly; is the response index of temperature fluctuation frequency; Calculating a cumulative thermal stress equivalent value based on the thermal stress equivalent value, wherein the calculation process of the cumulative thermal stress equivalent value includes calculating the number of effective sliding windows and calculating the cumulative thermal stress equivalent value; The number of effective sliding windows The calculation formula is: in, is the cumulative time period of thermal stress; is the time length of the sliding window; Based on the effective sliding window number Calculate the cumulative thermal stress equivalent value , the calculation formula is: in, For the Thermal stress equivalent value of each window; Based on the cumulative thermal stress equivalent value Generate the node constraint rules ,and ,in, is the maximum thermal stress threshold allowed by the thermal-sensitive chip material.

6. The production logistics planning system for the development of thermal chips based on temperature measurement according to claim 1, characterized in that: The thermal exposure history duration is the cumulative exposure time of the thermal-sensitive chip during the process of passing through all nodes of the path, and the tracking process of the thermal exposure history duration is the calculation of the effective thermal exposure time and the calculation of the thermal exposure history duration; For each node in the path , the temperature sequence is collected by the deployed temperature sensor , then the effective heat exposure time of the node is: ， in, For the The effective heat exposure time of each node, i.e., the transmission or residence time, in seconds (s); For nodes The total number of temperature sampling points; is an indicator variable; is the temperature sampling time interval; For the The temperature value of each sampling point; is the safety temperature threshold; Based on the effective heat exposure time The heat exposure history duration is calculated using the following formula: in, is the total duration of heat exposure history, in seconds; is the number of nodes traversed in the path.

7. A production logistics planning system for the development of thermal chips based on temperature measurement according to claim 6, characterized in that: The temperature change frequency data is the cumulative number of temperature fluctuations experienced by the thermal chip in the entire path, and the calculation formula is: in, is the total frequency of temperature fluctuations along the entire path, in times; For the The number of times the temperature sequence within a node crosses the mean temperature is used to indicate the severity of the temperature change; The calculation process of the path transport time cost includes calculating the path segment transport time and accumulating the total path segment transport time; The transport time of the path segment is composed of three parts: path length, transport speed and additional delay, and the calculation formula is: in, is the route segment transportation time; For path segments Length, in meters; is the average transport speed on the path segment, in meters per second; The system scheduling delay time is in seconds; Based on the route segment transit time The total transport time of the path segment is obtained by adding up , the calculation formula is: Based on the total duration of the heat exposure history The total frequency of temperature fluctuations along the entire path and the transport time of the route segment Get the optimal logistics path ， The formula is: Find in, is the optimal logistics path; is the path variable to be optimized, representing the set of nodes and path segments that the thermal chip passes through from the starting point to the end point; is a multi-objective function; Indicates that from all paths , find the path that minimizes the objective function.

8. The production logistics planning system for the development of thermal chips based on temperature measurement according to claim 5 is characterized in that: The cumulative thermal stress equivalent value based on the optimal logistics path The thermal stress risk coefficient ratio of the optimal logistics path is obtained: in, is the heat stress risk factor ratio, and ; The maximum thermal stress threshold allowed for the heat-sensitive chip material; The heat stress risk factor ratio and warning thresholds Compare to determine whether a response mechanism needs to be triggered; when When , it means that the cumulative thermal stress equivalent value is in the safe range and no response is required; when , it means that the accumulated thermal stress equivalent value is close to the material threshold and a response mechanism needs to be triggered.

9. The production logistics planning system for the development of thermal chips based on temperature measurement according to claim 1, characterized in that: The response mechanism is differentiated into whether the current path segment has the temperature control capability or the current path segment does not have the temperature control capability; When the current path segment has temperature control capability, the system activates a local temperature control device to reduce the current temperature. The calculation process of the thermal stress prediction value after temperature control includes calculating the expected reduced thermal stress equivalent value and generating the thermal stress prediction value after temperature control adjustment. The calculation formula for the expected reduction in thermal stress equivalent value is: in, To anticipate a reduction in thermal stress equivalent values; is the temperature control efficiency coefficient; is the current temperature of the path segment; is the target temperature control temperature; Based on the projected reduction in thermal stress equivalent values The thermal stress prediction value after the temperature control adjustment is calculated using the following formula: in, is the predicted value of thermal stress after temperature control adjustment; is the safety proportional factor; When the current path segment does not have the temperature control capability or the thermal stress prediction value still exceeds the standard after the temperature control adjustment, the system will call the dynamic path planning module M3, eliminate the current path segment from the total path segment and regenerate the optimal logistics path with the cumulative thermal stress equivalent value constraint condition.

Images