A special pressure gauge performance data monitoring and management system and method based on the Internet of Things

Abstract

The invention discloses a special pressure gauge performance data monitoring and management system and method based on the Internet of Things, which relates to the technical field of pressure gauge performance data monitoring, and includes a data acquisition module, a data storage module, a data analysis module, a pressure gauge performance test module and a control module; the data acquisition module is used to obtain the operating environment data of the pressure gauge; the data analysis module is used to establish a loss evaluation model for the pressure gauge; the data storage module stores operating environment data; the pressure gauge performance test is used to detect the performance of the pressure gauge; the control module is used to notify the management personnel to monitor the pressure gauge; the invention adjusts the loss by means of a simulated annealing algorithm; the loss is calculated by a multivariate linear regression model, and in the inner loop of the simulated annealing algorithm, the weight of the multivariate linear regression model is adjusted once, and when the simulated annealing algorithm ends, a global optimal solution or a better local optimal solution of the weight of the multivariate linear regression model is also obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of pressure gauge performance data monitoring, and in particular to a special pressure gauge performance data monitoring management system and method based on the Internet of Things. Background Art

[0002] A pressure gauge is an instrument for measuring pressure, usually used to measure the pressure of fluids or gases. A pressure gauge can measure pressure through a pressure sensor or a pressure transmitter, and convert the pressure signal into an electrical signal for measurement and recording. A pressure gauge is usually used in industrial production to monitor the pressure of fluids. The calibration cycle of a pressure gauge is usually one year or six months, depending on the use environment and requirements. During the calibration cycle, the pressure gauge needs to be calibrated and tested regularly to ensure that its performance meets the requirements. During the long-term use of the pressure gauge, the pressure gauge is affected by various environmental factors, such as high temperature, high humidity, strong vibration, etc. These factors may affect the measurement error of the pressure gauge. In addition, some acidic and alkaline substances may also affect the pressure gauge. Therefore, how to judge whether the pressure gauge needs to be calibrated based on the different environmental factors it is subject to has become a problem that needs to be solved. Summary of the invention

[0003] The purpose of the present invention is to provide a special pressure gauge performance data monitoring and management system and method based on the Internet of Things to solve the problems raised in the above background technology.

[0004] In one aspect of the present invention, a special pressure gauge performance data monitoring and management system based on the Internet of Things is provided, comprising: a data acquisition module, a data storage module, a data analysis module, a pressure gauge performance test module and a control module; the output end of the data acquisition module is interconnected with the input end of the data analysis module and the data storage module, and is used to obtain the operating environment data of the pressure gauge; the output end of the data analysis module is interconnected with the input end of the control module, and based on the operating environment data of the pressure gauge, a loss evaluation model of the pressure gauge is established to judge the loss of the pressure gauge; the data storage module is interconnected with the data analysis module, and is used to store the historical performance test data, operating environment data and product production information of the pressure gauge; the output end of the pressure gauge performance test module is interconnected with the input end of the data storage module, and is used to detect the performance of the pressure gauge; when the control module determines that the performance of the pressure gauge does not meet the requirements of the production specifications, it notifies the management personnel to monitor the pressure gauge and contacts the maintenance personnel to calibrate and maintain the pressure gauge.

[0005] Specifically, the data analysis module integrates the operating environment data of the pressure gauge to obtain the loss characteristics of the pressure gauge; takes the loss characteristics of the pressure gauge as input and the loss as output, trains a multivariate linear regression model as a loss evaluation model for the pressure gauge, and evaluates the performance of the pressure gauge; at the same time, uses a simulated annealing algorithm to adjust the loss, and then adjusts the weights of the multivariate linear regression model to optimize the effect of the pressure gauge performance evaluation model.

[0006] Specifically, the force gauge performance test module also includes a controller unit, a pressure source unit, a test data acquisition unit, a test data processing unit and a display unit; the controller unit is used to set test parameters and control the test process; the pressure source unit is used to test the pressure signal; the test data acquisition unit is used to collect the test data of the pressure gauge; the test data processing unit is used to analyze and process the collected pressure gauge test data, and calculate the performance indicators of the pressure gauge; the display unit is used to display the processed pressure gauge test results.

[0007] The data acquisition module includes a direct measurement unit and an indirect measurement unit. The direct measurement unit directly obtains environmental data that affects the pressure gauge through a sensor; the indirect measurement unit is used to extract environmental data that cannot be directly measured by a sensor from the production information of the product;

[0008] For example, in the process of industrial production, when a pressure gauge is used to measure the pressure of a fluid, the degree of corrosion cannot usually be measured directly. For industrial production, analyzing the concentration of the main product is a necessary process, and the concentration of the main product is correlated with the degree of corrosion of the fluid. In this case, the concentration of the main product can be used instead of the degree of corrosion of the fluid; changes in spring wire performance are one of the main sources of pressure gauge errors. Fluctuations in fluid pressure will affect the deformation of the spring wire, and thus affect the performance of the pressure gauge. By measuring the fluctuations in fluid pressure, the deformation data of the spring wire can be indirectly obtained, and measuring the fluctuations in fluid pressure is the purpose of installing a pressure gauge, which will not result in additional resource consumption.

[0009] In another aspect of the present invention, a method for monitoring and managing performance data of a special pressure gauge based on the Internet of Things is provided, comprising the following steps:

[0010] S5-1, obtaining usage data of the pressure gauge, and extracting a loss characteristic from the usage data of the pressure gauge;

[0011] S5-2, based on the usage data of the pressure gauge, establish a pressure gauge loss assessment model;

[0012] S5-3, using the pressure gauge loss assessment model to determine whether the measurement error of the current pressure gauge meets the requirements of the production specifications, if so, repeat step S5-3; if not, identify and repair the pressure gauge.

[0013] In step S5-2, the pressure gauge loss evaluation model is established based on the loss characteristics of the pressure gauge, and the following steps are also included:

[0014] The loss characteristic of the pressure gauge is X 1 , X 2 , …, X n , n is the number of loss characteristics; taking the loss characteristics of the pressure gauge as input and the loss of the pressure gauge as output, the n-element linear regression model is trained as the pressure gauge loss evaluation model. , where i is a positive integer between [1, m], m is the number of historical usage data of the pressure gauge; k is a natural number, indicating the number of iterations; , , …, Represents the weight of the loss feature after the kth iteration; , , …, Represents the loss characteristics of the historical usage data of the i-th pressure gauge; Represents the regression value of the loss degree of the i-th pressure gauge after the k-th iteration; optimizes the weights of the n-variable linear regression model through the simulated annealing algorithm;

[0015] Different environmental factors are connected through a multiple linear regression model to obtain the overall loss of the pressure gauge; the measurement error of the pressure gauge should be positively correlated with the loss, so it is necessary to adjust the loss of the pressure gauge according to the measurement error of the pressure gauge, and then optimize the weight of the multiple linear regression model; the measurement error of the pressure gauge is determined by testing. Before using the method provided by the present invention, the pressure gauge is usually calibrated at a period of one year or half a year, and the measurement error of the pressure gauge can be obtained from the historical calibration results.

[0016] Optimizing the weights of the n-variable linear regression model by simulated annealing algorithm specifically includes the following steps:

[0017] S7-1, set the initial temperature T 0 , set the loss feature weight , , …, =1, the loss feature , , …, Substitute into the formula to calculate the initial value of the loss , , set the consumption feature weight , , …, and the initial value of the loss As the initial solution, determine the loss value loss corresponding to the initial solution 0 ; Set the initial solution as the current solution, the initial temperature as the current temperature, and set the initial value of the iteration number k to 0;

[0018] In the historical usage data of the pressure gauge, find the historical usage data of the pressure gauge whose measurement error is closest to the threshold, and set the initial loss value of the historical usage data of the pressure gauge to As the decision boundary, , compare the initial loss value of other pressure gauge historical usage data with the decision boundary ; If the initial value of the loss degree is greater than or equal to the decision boundary, it is judged that the pressure gauge needs to be overhauled; otherwise, the pressure gauge does not need to be overhauled; the obtained decision accuracy is used as the loss value; the correct decision includes two cases: the regression value of the pressure gauge loss degree is greater than or equal to the decision boundary and the measurement error of the pressure gauge is greater than or equal to the requirements of the production specifications, and the regression value of the pressure gauge loss degree is less than the decision boundary and the measurement error of the pressure gauge is less than the requirements of the production specifications; the initial value of the loss degree is the same as the regression value of the loss degree, because the initial value of the loss degree is the regression value when the weights are all 1, and the decision accuracy can be judged directly by using the loss degree; starting from the next iteration, the loss degree and the loss degree regression value are different, and the decision accuracy is calculated using the loss degree regression value;

[0019] S7-2, for counting units b=1, 2, ..., L, repeat steps S7-3 to S7-4; L is the number of inner loops;

[0020] S7-3, adding 1 to the value of the iteration number k, changing the loss feature weight and loss by generating a disturbance based on the current solution, taking the loss feature weight and loss selected after the disturbance as a new solution, and determining the decision accuracy p corresponding to the new solution;

[0021] S7-4, calculate the increment Δp of the expected value of the response time brought by the new solution. If the increment is less than 0, the new solution is accepted as the new current solution with probability 1. If the increment is greater than or equal to 0, the new solution is accepted as the new current solution with probability Accept the new solution as the new current solution;

[0022] If the new solution is accepted, k remains unchanged; if the new solution is not accepted, the value of k decreases by 1, and the parameters of the n-variable linear regression model do not change;

[0023] S7-5, select a cooling plan, and lower the current temperature according to the cooling plan. If the current temperature is greater than or equal to the set threshold, enter step S7-2; if the current temperature is less than the set threshold, determine the loss feature weight and loss degree based on the current solution.

[0024] Specifically, the loss characteristics , , …, The calculation is done by the following formula:

[0025] , , …, ; In the formula , , …, is the measured value of the loss characteristic of the historical usage data of the i-th pressure gauge at time t, where t is the integral variable;

[0026] The integral interval is the time period between two calibrations of the pressure gauge. Due to the limitation of equipment conditions, the measurement of the loss characteristic is discrete. The loss characteristic can be approximately calculated: ; In the formula The time point between two calibrations of the pressure gauge. To measure the measuring period of the device, the area of ​​the curved trapezoid is approximated by the sum of the areas of multiple rectangles; the remaining loss characteristics are approximated in a similar manner;

[0027] In step S7-3, the perturbation is generated based on the current solution, and the loss feature weight and loss are changed, which further includes the following steps:

[0028] S9-1, in the current solution , , …, and Based on the probability, the loss of all pressure gauges is adjusted to obtain ;

[0029] S9-2, the loss characteristic , , …, As input, the loss of the pressure gauge As output, train the n-ary linear regression model to obtain the loss feature weights , , …, ;

[0030] The input of the n-variable linear regression model remains unchanged, and the output loss degree is adjusted with the iteration. After the loss degree is adjusted, the weight of the n-variable linear regression model needs to be adjusted, and the regression value of the historical usage data of the pressure gauge is calculated using the adjusted weight. The regression value is compared with the decision boundary to obtain the decision accuracy, which is used to determine whether it is necessary to accept the iteratively adjusted loss degree;

[0031] S9-4, in the historical usage data of the pressure gauge, find the loss characteristic of the historical usage data of the pressure gauge whose measurement error is closest to the threshold , , …, , , according to the loss feature weight , , …, Calculate the loss of the pressure gauge based on historical usage data , as the new decision boundary; calculate the loss regression value of the historical usage data of other pressure gauges to obtain ;

[0032] The present invention uses a linear regression model to determine whether the current pressure gauge needs to be calibrated, inputs the loss characteristics of the current pressure gauge into the linear regression model to obtain a regression value of the loss; and compares the regression value of the loss with the decision boundary to determine whether calibration is required.

[0033] The decision accuracy is determined by:

[0034] Compare the loss regression values ​​of the pressure gauge's historical usage data With decision boundary , count the number of historical usage data of the pressure gauge with correct decisions NUM1, take the ratio of NUM1 to m NUM1 / m as the decision accuracy rate, and the historical usage data data does not participate in the calculation;

[0035] Adjust the loss of all pressure gauges according to probability to obtain The following steps are involved:

[0036] S10-1, mis 1 、mis 2 、…、mis m represents the historical measurement error of the pressure gauge, then mis max is the measurement error closest to the threshold; calculate mis 1 、mis 2 、…、mis m with mis max The algebraic difference of 1 , Δmis 2 ,…,Δmism ;

[0037] The measurement error is always a positive number. No matter whether the measured value is greater than or less than the actual value, the absolute value is taken as the measurement error.

[0038] S10-2, for i=1, 2, ...m, execute step S10-3;

[0039] S10-3, by random number according to probability Will Adjust to , , is a symbolic variable, if is greater than or equal to the threshold, and is less than the decision boundary, then is 1; if is greater than or equal to the threshold, and If is greater than the decision boundary, is 0; if is less than the threshold, and is less than the decision boundary, then is 0; if is less than the threshold, and If is greater than the decision boundary, is 1; and The absolute value of is positively correlated;

[0040] Algebraic Difference With positive and negative signs, when When it is less than 0, it means that the performance of the pressure gauge meets the requirements of the production rules, and the loss should be adjusted in the direction of reduction; when When it is greater than 0, it means that the performance of the pressure gauge does not meet the requirements of the production rules, and the loss degree should be adjusted in the direction of increase;

[0041] For the pressure gauge usage data that has made correct decisions, it is not necessary to adjust the loss degree. Under the current decision boundary, the decision accuracy cannot be improved, so the symbol variable is set to 0; for the pressure gauge usage data that has not made correct decisions, adjusting all of them may not be optimal, and the best adjustment method needs to be found, so the loss degree of these usage data is iteratively adjusted according to probability through the simulated annealing algorithm.

[0042] Compared with the prior art, the beneficial effects achieved by the present invention are: adjusting the loss degree with the help of a simulated annealing algorithm, accepting a poor loss degree combination with a certain probability, which is conducive to avoiding falling into a local optimal solution and finding a global optimal solution; calculating the loss degree with a multivariate linear regression model, in the inner loop of the simulated annealing algorithm, adjusting the weights of the multivariate linear regression model once for each step executed, and at the end of the simulated annealing algorithm, the global optimal solution or a better local optimal solution of the weights of the multivariate linear regression model is also obtained; indirectly measuring the environmental factors that affect the pressure gauge helps to reduce resource expenditure and improve the profit margin of the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0044] Figure 1 It is a structural schematic diagram of a special pressure gauge performance data monitoring and management system based on the Internet of Things according to an embodiment of the present invention. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] In the embodiments of the present invention, please refer to Figure 1 , provides a special pressure gauge performance data monitoring and management system based on the Internet of Things, including: a data acquisition module, a data storage module, a data analysis module, a pressure gauge performance test module and a control module; the output end of the data acquisition module is interconnected with the input end of the data analysis module and the data storage module to obtain the operating environment data of the pressure gauge; the output end of the data analysis module is interconnected with the input end of the control module, and based on the operating environment data of the pressure gauge, a loss evaluation model of the pressure gauge is established to judge the loss of the pressure gauge; the data storage module is interconnected with the data analysis module to store the historical performance test data, operating environment data and production information of the pressure gauge; the output end of the pressure gauge performance test module is interconnected with the input end of the data storage module to detect the performance of the pressure gauge; the control module, when determining that the performance of the pressure gauge does not meet the requirements of the production specifications, notifies the management personnel to monitor the pressure gauge, and contacts the maintenance personnel to calibrate and maintain the pressure gauge.

[0047] The data analysis module integrates the operating environment data of the pressure gauge to obtain the loss characteristics of the pressure gauge; uses the loss characteristics of the pressure gauge as input and the loss as output to train a multivariate linear regression model as a loss evaluation model for the pressure gauge to evaluate the performance of the pressure gauge; at the same time, uses a simulated annealing algorithm to adjust the loss, and then adjusts the weights of the multivariate linear regression model to optimize the effect of the pressure gauge performance evaluation model.

[0048] The pressure gauge performance test module also includes a controller unit, a pressure source unit, a test data acquisition unit, a test data processing unit and a display unit; the controller unit is used to set test parameters and control the test process; the pressure source unit is used to test the pressure signal; the test data acquisition unit is used to collect the test data of the pressure gauge; the test data processing unit is used to analyze and process the collected pressure gauge test data and calculate the performance indicators of the pressure gauge; the display unit is used to display the processed pressure gauge test results.

[0049] The data acquisition module includes a direct measurement unit and an indirect measurement unit. The direct measurement unit directly obtains environmental data that affects the pressure gauge through a sensor; the indirect measurement unit is used to extract environmental data that cannot be directly measured by a sensor from the production information of the product.

[0050] In an embodiment of the present invention, a method for monitoring and managing performance data of a special pressure gauge based on the Internet of Things is provided, comprising the following steps:

[0051] S5-1, obtaining usage data of the pressure gauge, and extracting a loss characteristic from the usage data of the pressure gauge;

[0052] The loss characteristic of the pressure gauge is X 1 , X 2 , X 3 , X 4 , respectively represent relative humidity, temperature, corrosivity and pressure change rate; among which the corrosivity is indirectly expressed by detecting the concentration of the main product in the production process;

[0053] S5-2, based on the usage data of the pressure gauge, establish a pressure gauge loss assessment model;

[0054] With the loss characteristics of the pressure gauge as input and the loss degree of the pressure gauge as output, a 4-element linear regression model is trained as the pressure gauge loss evaluation model. , where i is a positive integer between [1, m], m is the number of historical usage data of the pressure gauge; k is a natural number, indicating the number of iterations; , , , Represents the weight of the loss feature after the kth iteration; , , , Represents the loss characteristics of the historical usage data of the i-th pressure gauge; It represents the regression value of the loss degree of the i-th pressure gauge after the k-th iteration;

[0055] Loss characteristics , , , The calculation is done by the following formula:

[0056] , , …, ; In the formula , , …, is the measured value of the loss characteristic of the historical usage data of the i-th pressure gauge at time t, where t is the integral variable;

[0057] The weights of the 4-variable linear regression model are optimized by a simulated annealing algorithm, including the following steps S7-1 to S7-5:

[0058] S7-1, set the initial temperature T 0 Set the loss feature weight to 100 , , , =1, the loss feature , , , Substitute into the formula to calculate the initial value of the loss , , set the consumption feature weight , , , and the initial value of the loss As the initial solution, determine the loss value loss corresponding to the initial solution 0 ; Set the initial solution as the current solution, the initial temperature as the current temperature, and set the initial value of the iteration number k to 0;

[0059] In the historical usage data of the pressure gauge, find the historical usage data of the pressure gauge whose measurement error is closest to the threshold, and set the initial loss value of the historical usage data of the pressure gauge to As the decision boundary, , compare the initial loss value of other pressure gauge historical usage data with the decision boundary ; If the initial value of the loss degree is greater than or equal to the decision boundary, it is judged that the pressure gauge needs to be repaired; otherwise, the pressure gauge does not need to be repaired; the obtained decision accuracy is used as the loss value; the correct decision includes two situations: the regression value of the pressure gauge loss degree is greater than or equal to the decision boundary and the measurement error of the pressure gauge is greater than or equal to the requirements of the production specifications, and the regression value of the pressure gauge loss degree is less than the decision boundary and the measurement error of the pressure gauge is less than the requirements of the production specifications;

[0060] S7-2, for counting unit b=1, 2, ..., 20, repeat steps S7-3 to S7-4;

[0061] S7-3, add 1 to the value of the iteration number k, change the loss characteristic weight and loss by generating disturbance on the basis of the current solution, take the loss characteristic weight and loss selected after the disturbance as the new solution, and determine the decision accuracy p corresponding to the new solution; compare the loss regression value of the historical usage data of the pressure gauge With decision boundary , count the number of historical usage data of the pressure gauge with correct decisions NUM1, take the ratio of NUM1 to m NUM1 / m as the decision accuracy rate, and the historical usage data of the pressure gauge data does not participate in the calculation;

[0062] Generating a disturbance based on the current solution includes the following steps S9-1 to S9-4:

[0063] S9-1, in the current solution , , , and Based on the probability, the loss of all pressure gauges is adjusted to obtain ; including steps S10-1 to S10-3,

[0064] S10-1, mis 1 、mis 2 、…、mis m represents the historical measurement error of the pressure gauge, then mis max is the measurement error closest to the threshold; calculate mis 1 、mis 2 、…、mis m with mis max The algebraic difference of 1 , Δmis 2 ,…,Δmis m ;

[0065] S10-2, for i=1, 2, ...m, execute step S10-3;

[0066] S10-3, by random number according to probability Will Adjust to , , is a symbolic variable, if is greater than or equal to the threshold, and is less than the decision boundary, then is 1; if is greater than or equal to the threshold, and If is greater than the decision boundary, is 0; if is less than the threshold, and is less than the decision boundary, then is 0; if is less than the threshold, and If is greater than the decision boundary, is 1; and The absolute value of is positively correlated; G , G is a positive constant;

[0067] S9-2, the loss characteristic , , , As input, the loss of the pressure gauge As output, train a 4-tuple linear regression model to obtain the loss feature weights , , , ;

[0068] S9-4, in the historical usage data of the pressure gauge, find the loss characteristic of the historical usage data of the pressure gauge whose measurement error is closest to the threshold , , , , , according to the loss feature weight , , , Calculate the loss of the pressure gauge based on historical usage data , as the new decision boundary; calculate the loss regression value of the historical usage data of other pressure gauges to obtain ;

[0069] S7-4, calculate the increment Δp of the expected value of the response time brought by the new solution. If the increment is less than 0, the new solution is accepted as the new current solution with probability 1. If the increment is greater than or equal to 0, the new solution is accepted as the new current solution with probability Accept the new solution as the new current solution;

[0070] If the new solution is accepted, k remains unchanged; if the new solution is not accepted, the value of k decreases by 1, and the parameters of the 4-variable linear regression model do not change;

[0071] S7-5, select a cooling scheme, and lower the current temperature according to the cooling scheme. If the current temperature is greater than or equal to the set threshold, proceed to step S7-2; if the current temperature is less than the set threshold, determine the loss characteristic weight and loss according to the current solution; when a linear cooling curve with a cooling coefficient of 0.9 is used for cooling, the temperature changes from 100°C to 100×0.9°C after the first iteration; the threshold is set to a smaller value, here the threshold can be taken as 1°C

[0072] The specific process of updating the weights of the 4-element linear regression model is as follows:

[0073] First, calculate the historical usage data of m pressure gauges , , , , calculate the loss of each data according to the formula , , at this time, the loss degree and the loss degree regression value Similarly, find the loss degree of the historical usage data of the pressure gauge whose measurement error is closest to the threshold , as the decision boundary, the measurement error is determined by the national regulations or the requirements of the generated specifications, and the decision accuracy is calculated based on the decision boundary and the loss regression value;

[0074] For the first iteration, Adjust to ,by , , , For input, As output, we get the weights of the 4-element linear regression model , , , , according to the weight , , , and input , , , Get the loss regression value , the new decision boundary changes to , according to step S10-3, and Same, will not change, manually Also changed to , the decision accuracy is calculated based on the decision boundary and loss regression value, and whether to accept it is determined according to the probability , , , and ; If accepted, add 1 to the number of iterations k. , , , , Will be in , , , , If not accepted, k remains unchanged and continues to , , , , Produced on the basis of , , , , ; Continue to repeat the above steps to adjust the weights of the 4-element linear regression model until the simulated annealing algorithm ends;

[0075] S5-3, using the pressure gauge loss assessment model to determine whether the measurement error of the current pressure gauge meets the requirements of the production specifications, if so, repeat step S5-3; if not, identify and repair the pressure gauge.

[0076] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0077] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A special pressure gauge performance data monitoring and management system based on the Internet of Things, characterized in that: It includes a data acquisition module, a data storage module, a data analysis module, a pressure gauge performance test module and a control module; the output end of the data acquisition module is interconnected with the input end of the data analysis module and the data storage module to obtain the operating environment data of the pressure gauge; the output end of the data analysis module is interconnected with the input end of the control module, and based on the operating environment data of the pressure gauge, a loss evaluation model of the pressure gauge is established to determine the loss of the pressure gauge; the data storage module is interconnected with the data analysis module to store the historical performance test data of the pressure gauge, the operating environment data and the production information of the product; The output end of the pressure gauge performance test module is connected to the input end of the data storage module to detect the performance of the pressure gauge; The control module, when determining that the performance of the pressure gauge does not meet the requirements of the production specifications, notifies the management personnel to monitor the pressure gauge and contacts the maintenance personnel to calibrate and maintain the pressure gauge; The data analysis module obtains the loss characteristic of the pressure gauge by integrating the operating environment data of the pressure gauge; The loss characteristics of the pressure gauge are taken as input and the loss as output. A multivariate linear regression model is trained as the loss evaluation model of the pressure gauge to evaluate the performance of the pressure gauge. At the same time, the simulated annealing algorithm is used to adjust the loss, and then the weights of the multivariate linear regression model are adjusted to optimize the effect of the pressure gauge performance evaluation model.

2. According to the Internet of Things-based special pressure gauge performance data monitoring and management system of claim 1, it is characterized in that: The pressure gauge performance test module also includes a controller unit, a pressure source unit, a test data acquisition unit, a test data processing unit and a display unit; the controller unit is used to set test parameters and control the test process; the pressure source unit is used for the test pressure signal; the test data acquisition unit is used to collect the test data of the pressure gauge; The test data processing unit is used to analyze and process the collected pressure gauge test data and calculate the performance index of the pressure gauge; the display unit is used to display the processed pressure gauge test results.

3. According to the Internet of Things-based special pressure gauge performance data monitoring and management system of claim 1, it is characterized in that: The data acquisition module includes a direct measurement unit and an indirect measurement unit. The direct measurement unit directly obtains environmental data that affects the pressure gauge through a sensor; the indirect measurement unit is used to extract environmental data that cannot be directly measured by a sensor from the production information of the product.

4. A special pressure gauge performance data monitoring and management method based on the Internet of Things, characterized in that: The following steps are involved: S5-1, obtaining usage data of the pressure gauge, and extracting a loss characteristic from the usage data of the pressure gauge; S5-2, based on the usage data of the pressure gauge, establish a pressure gauge loss evaluation model; obtain the loss characteristics of the pressure gauge, take the loss characteristics of the pressure gauge as input and the loss of the pressure gauge as output, train a linear regression model as the pressure gauge loss evaluation model; and optimize the weights of the linear regression model through a simulated annealing algorithm; S5-3, using the pressure gauge loss assessment model to determine whether the measurement error of the current pressure gauge meets the requirements of the production specifications, if so, repeat step S5-3; if not, identify and repair the pressure gauge.

5. The method for monitoring and managing performance data of special pressure gauges based on the Internet of Things according to claim 4 is characterized in that: In step S5-2, the pressure gauge loss evaluation model is established based on the loss characteristics of the pressure gauge, and the following steps are also included: The loss characteristics of the pressure gauge are X1, X2, ..., X n , n is the number of loss characteristics; taking the loss characteristics of the pressure gauge as input and the loss of the pressure gauge as output, the n-element linear regression model is trained as the pressure gauge loss evaluation model. , where i is a positive integer between [1, m], m is the number of historical usage data of the pressure gauge; k is a natural number, indicating the number of iterations; , , …, Represents the weight of the loss feature after the kth iteration; , , …, Represents the loss characteristics of the historical usage data of the i-th pressure gauge; Represents the regression value of the loss degree of the i-th pressure gauge after the k-th iteration; the weights of the n-variable linear regression model are optimized by the simulated annealing algorithm.

6. A method for monitoring and managing performance data of special pressure gauges based on the Internet of Things according to claim 5, characterized in that: The method of optimizing the weights of the n-variable linear regression model by using a simulated annealing algorithm specifically includes the following steps: S7-1, set the initial temperature T0, set the loss characteristic weight , , …, =1, the loss feature , , …, Substitute into the formula to calculate the initial value of the loss , , set the consumption feature weight , , …, and the initial value of the loss As the initial solution, determine the loss value loss0 corresponding to the initial solution; take the initial solution as the current solution, take the initial temperature as the current temperature, and set the initial value of the number of iterations k to 0; In the historical usage data of the pressure gauge, find the historical usage data of the pressure gauge whose measurement error is closest to the threshold, and set the initial loss value of the historical usage data of the pressure gauge to As the decision boundary, , compare the initial loss value of other pressure gauge historical usage data with the decision boundary ; If the initial value of the loss degree is greater than or equal to the decision boundary, it is judged that the pressure gauge needs to be repaired; otherwise, the pressure gauge does not need to be repaired; the obtained decision accuracy is used as the loss value; the correct decision includes two situations: the regression value of the pressure gauge loss degree is greater than or equal to the decision boundary and the measurement error of the pressure gauge is greater than or equal to the requirements of the production specifications, and the regression value of the pressure gauge loss degree is less than the decision boundary and the measurement error of the pressure gauge is less than the requirements of the production specifications; S7-2, for counting units b=1, 2, ..., L, repeat steps S7-3 to S7-4; L is the number of inner loops; S7-3, adding 1 to the value of the iteration number k, changing the loss feature weight and loss by generating a disturbance based on the current solution, taking the loss feature weight and loss selected after the disturbance as a new solution, and determining the decision accuracy p corresponding to the new solution; S7-4, calculate the increment Δp of the expected value of the response time brought by the new solution. If the increment is less than 0, the new solution is accepted as the new current solution with probability 1. If the increment is greater than or equal to 0, the new solution is accepted as the new current solution with probability Accept the new solution as the new current solution; If the new solution is accepted, k remains unchanged; if the new solution is not accepted, the value of k decreases by 1, and the parameters of the n-variable linear regression model do not change; S7-5, select a cooling plan, and lower the current temperature according to the cooling plan. If the current temperature is greater than or equal to the set threshold, enter step S7-2; if the current temperature is less than the set threshold, determine the loss feature weight and loss degree based on the current solution.

7. The method for monitoring and managing performance data of special pressure gauges based on the Internet of Things according to claim 5 is characterized in that: Loss characteristics , , …, The calculation is done by the following formula: , , …, ; In the formula , , …, is the measured value of the loss characteristic of the historical usage data of the ith pressure gauge at time t, and t is the integral variable.

8. The method for monitoring and managing performance data of special pressure gauges based on the Internet of Things according to claim 7 is characterized in that: In step S7-3, the perturbation is generated based on the current solution, and the loss feature weight and loss are changed, which further includes the following steps: S9-1, in the current solution , , …, and Based on the probability, the loss of all pressure gauges is adjusted to obtain ; S9-2, the loss characteristic , , …, As input, the loss of the pressure gauge As output, train the n-ary linear regression model to obtain the loss feature weights , , …, ; S9-4, in the historical usage data of the pressure gauge, find the loss characteristic of the historical usage data of the pressure gauge whose measurement error is closest to the threshold , , …, , , according to the loss feature weight , , …, Calculate the loss of the pressure gauge based on historical usage data , as the new decision boundary; calculate the loss regression value of the historical usage data of other pressure gauges to obtain .

9. A method for monitoring and managing performance data of special pressure gauges based on the Internet of Things according to claim 8, characterized in that: The decision accuracy is determined by: Compare the loss regression values ​​of the pressure gauge's historical usage data With decision boundary , count the number of historical usage data of the pressure gauge with correct decisions NUM1, take the ratio of NUM1 to m NUM1 / m as the decision accuracy rate, and the historical usage data of the pressure gauge data does not participate in the calculation; Adjust the loss of all pressure gauges according to probability to obtain The following steps are involved: S10-1, mis1, mis2, ..., mis m represents the historical measurement error of the pressure gauge, then mis max is the measurement error closest to the threshold; calculate mis1, mis2, ..., mis m with mis max The algebraic difference is Δmis1, Δmis2, …, Δmis m ; S10-2, for i=1, 2, ...m, execute step S10-3; S10-3, by random number according to probability Will Adjust to , , is a symbolic variable, if is greater than or equal to the threshold, and is less than the decision boundary, then is 1; if is greater than or equal to the threshold, and is greater than the decision boundary, then is 0; if is less than the threshold, and is less than the decision boundary, then is 0; if is less than the threshold, and is greater than the decision boundary, then is 1; and The absolute value of is positively correlated.

Images