A power plant production management system

Abstract

The present application relates to the technical field of device management, in particular to a power plant production management system, the present application utilizes multiple sensors and deep learning algorithm, can realize power plant device management well, the algorithm of the present application can according to the equipment operating state and corresponding maintenance and repair condition involved in production system, give the difficulty expectation of equipment maintenance and repair, and the expected maintenance interval, the present application can let maintenance and repair personnel, according to algorithm formula make specific maintenance maintenance difficulty evaluation and corresponding repair interval, so as to make advance planning to equipment update, make the expectation of expected replacement maintenance of equipment, regularly update equipment maintenance information, guarantee the stability of system operation.

Description

Technical Field

[0001] This invention relates to the field of equipment management technology, specifically a power plant production management system. Background Technology

[0002] Any malfunction in the power plant's production process not only directly affects the output and quality of electricity, but may also cause equipment damage and personal injury accidents. To ensure that production equipment, operating equipment, and power generation equipment can operate safely, reliably, and effectively, and to enable them to play their full role, regular maintenance and repair, as well as early maintenance of the equipment, are essential.

[0003] The power generation process in a power plant involves the management of multiple pieces of equipment, requiring regular maintenance or repair by personnel. However, based on human judgment or equipment operation manuals, environmental issues or equipment malfunctions can lead to premature equipment failures or even damage. Therefore, this problem needs to be addressed. Summary of the Invention

[0004] The purpose of this invention is to provide a power plant production management system to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A power plant production management system includes the following steps: an information collection module, a model processing module, a data correction module, and an information analysis module.

[0007] The information collection module is a process of acquiring data based on ambient temperature sensors, ambient humidity sensors, electrical control system load sensors, current sensors, and temperature sensors.

[0008] Then, the model processing module can build a model based on the temperature, humidity and load intensity data of the equipment; using the equipment maintenance and repair demand data as labels, a prediction model for equipment maintenance and repair demand is established based on deep learning algorithms;

[0009] The data correction module can correct the predicted equipment maintenance and repair difficulty data based on the collected electrical variable data and the internal operating temperature data of the equipment, and obtain a comprehensive equipment maintenance and repair demand index.

[0010] Based on the obtained comprehensive equipment maintenance and repair demand index, the information analysis module uses a dynamic adjustment formula to calculate the equipment maintenance and repair interval. According to the calculated maintenance and repair interval, the system will provide an adjustment time for maintenance and repair. When the comprehensive maintenance and repair demand index is higher than the preset maintenance and repair demand threshold, the system will increase the maintenance difficulty indicator.

[0011] Step 1: Collect the ambient temperature of the power equipment using an ambient temperature sensor; collect the ambient humidity of the power equipment using an ambient humidity sensor; collect the load intensity of the equipment using an electrical control system load sensor; measure the electrical variables in the equipment's operating status using a current sensor; collect the internal operating temperature of the equipment and data on the difficulty of equipment maintenance and repair using a temperature sensor.

[0012] Step 2: Use the temperature, humidity and load intensity data of the equipment as the training set, and the corresponding equipment maintenance and repair demand data as the label. Build an equipment maintenance and repair demand prediction model based on deep learning algorithm, train the model, and obtain a trained equipment maintenance and repair demand difficulty prediction model.

[0013] Step 3: Use sensors to collect data on ambient temperature, humidity and load intensity of the equipment, and input them into the equipment maintenance and repair demand difficulty data prediction model to obtain the predicted equipment maintenance and repair demand data output by the model. Use the collected electrical variable data and the internal working temperature data of the equipment to correct the predicted equipment maintenance and repair demand difficulty data to obtain the comprehensive equipment maintenance and repair demand index.

[0014] Step 4: Based on the comprehensive equipment maintenance and repair demand index obtained above, the equipment maintenance and repair interval is calculated using a dynamic adjustment formula. According to the calculated maintenance and repair interval, the system will provide an adjustment time for maintenance and repair. When the comprehensive maintenance and repair demand index is higher than the preset maintenance and repair demand threshold, the system will increase the maintenance difficulty indicator.

[0015] Ambient temperature and humidity sensors are installed in different locations in the environment where the equipment is located. Load sensors, current sensors, and temperature sensors in the electrical control system are installed inside the equipment to collect corresponding data. Multiple sets of ambient temperature and humidity sensors can be installed and the average value is used as the data to be recorded. Ambient temperature and humidity sensors should be kept away from the equipment to avoid the temperature and moisture generated by the equipment itself affecting the accuracy of the data.

[0016] Equipment maintenance and repair requirements data are obtained based on historical equipment maintenance and repair data;

[0017] The load intensity data of the equipment is divided into the initial operation stage, the normal operation stage, and the tail operation stage.

[0018] The data output from each sensor is in a unified format, including timestamp, sensor type, and reading value, and the data will be securely backed up to a medium.

[0019] The data from each sensor is preprocessed to ensure data quality. The specific steps are as follows:

[0020] Remove outliers: Calculate the standard deviation of the data and set 3 times the standard deviation as a threshold. Data points that exceed this range are considered outliers.

[0021] The formula used to calculate the standard deviation of the data is:

[0022]

[0023] in, This is the standard deviation, and N is the total number of data points. It is the first Data points, It is the average value of the data;

[0024] The formula used to calculate the average of the data is:

[0025]

[0026] in, N is the average of the data, and N is the total number of data points. It is the first One data point;

[0027] Data interpolation: For time series data, linear interpolation is used to fill in missing values: assuming that the data changes linearly between two known points, missing values ​​are estimated by calculating the linear relationship between these two points.

[0028] The process of establishing a predictive model for equipment maintenance and repair needs based on deep learning algorithms specifically includes:

[0029] A prediction model for equipment maintenance and repair needs is established based on deep learning algorithms. The model consists of an input layer, a hidden layer, an activation layer, and an output layer. The temperature, humidity, and load intensity data of the equipment are used as the training set, and the corresponding equipment maintenance and repair needs data are used as labels to train the model, resulting in a well-trained prediction model for equipment maintenance and repair needs.

[0030] The predicted equipment maintenance and repair demand data L is corrected using the collected electrical variable data and the equipment's internal operating temperature data, resulting in the following formula for the comprehensive equipment maintenance and repair demand index:

[0031]

[0032] Where X is the comprehensive equipment maintenance and repair demand index, S is electrical variable data, L is the equipment internal operating temperature data, and H is the predicted equipment maintenance and repair difficulty data. It is the preset scaling factor for electrical variable data. It is a preset proportional coefficient for the internal operating temperature data of the equipment. It is a preset proportional coefficient for the predicted difficulty data of equipment maintenance and repair needs. , , Both are greater than zero, and > > , It is a constant correction exponent.

[0033] The maintenance and repair demand index for comprehensive equipment is used to calculate the maintenance and repair interval using a dynamic adjustment formula, which is as follows:

[0034]

[0035] in, It is the calculated maintenance and repair interval. This is the initial maintenance interval, in days, where X is the comprehensive equipment maintenance demand index. It is a preset adjustment coefficient, and Greater than zero.

[0036] Based on the calculated maintenance intervals, the system will automatically adjust the difficulty of equipment maintenance requirements, specifically including:

[0037] When the comprehensive equipment maintenance and repair demand index X is higher than the preset water demand threshold That is, X > At this time, the system will increase the difficulty of equipment maintenance and repair needs. The formula for calculating the maintenance and repair interval is as follows:

[0038]

[0039] in, The difficulty of equipment maintenance and repair needs X represents the basic maintenance difficulty, while X represents the comprehensive equipment maintenance and repair demand index. This is the preset proportional coefficient for the comprehensive equipment maintenance and repair demand index X. Greater than zero.

[0040] Preferably, the load sensor and current sensor of the electrical control system are installed on the main circuit of the equipment to monitor the working status of the equipment. The load sensor and current sensor of the electrical control system are connected in parallel in the circuit, do not interfere with each other, and are used to collect the equipment operation information at the same time.

[0041] Compared with the prior art, the beneficial effects of the present invention are:

[0042] This invention utilizes multiple sensors and deep learning algorithms to effectively manage power plant equipment. The algorithm can predict the difficulty of equipment maintenance and repair, as well as the expected maintenance intervals, based on the operating status and maintenance conditions of the equipment involved in the production system. This invention allows maintenance personnel to make specific assessments of maintenance difficulty and corresponding repair intervals based on the algorithm formula. This enables advance planning for equipment upgrades, anticipation of expected equipment replacement and repair, and regular updates to equipment maintenance information, ensuring stable system operation. Attached Figure Description

[0043] Figure 1 This is a schematic diagram of the process of the present invention.

[0044] Figure 2 This is a schematic diagram of the modules of the present invention. Detailed Implementation

[0045] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0046] Please see Figures 1 to 2 The present invention provides a technical solution:

[0047] A power plant production management system includes the following steps: an information collection module, a model processing module, a data correction module, and an information analysis module.

[0048] The information collection module is a process of acquiring data based on ambient temperature sensors, ambient humidity sensors, electrical control system load sensors, current sensors, and temperature sensors.

[0049] Then, the model processing module can build a model based on the temperature, humidity and load intensity data of the equipment; using the equipment maintenance and repair demand data as labels, a prediction model for equipment maintenance and repair demand is established based on deep learning algorithms;

[0050] The data correction module can correct the predicted equipment maintenance and repair difficulty data based on the collected electrical variable data and the internal operating temperature data of the equipment, and obtain a comprehensive equipment maintenance and repair demand index.

[0051] Based on the obtained comprehensive equipment maintenance and repair demand index, the information analysis module uses a dynamic adjustment formula to calculate the equipment maintenance and repair interval. According to the calculated maintenance and repair interval, the system will provide an adjustment time for maintenance and repair. When the comprehensive maintenance and repair demand index is higher than the preset maintenance and repair demand threshold, the system will increase the maintenance difficulty indicator.

[0052] Step 1: Collect the ambient temperature of the power equipment using an ambient temperature sensor; collect the ambient humidity of the power equipment using an ambient humidity sensor; collect the load intensity of the equipment using an electrical control system load sensor; measure the electrical variables in the equipment's operating status using a current sensor; collect the internal operating temperature of the equipment and data on the difficulty of equipment maintenance and repair using a temperature sensor.

[0053] In the embodiments of the present invention, multiple sets of current ambient temperature sensors and ambient humidity sensors need to be set up and placed far away from the equipment in order to collect more accurate environmental information and avoid the actual judgment being affected by the temperature of the equipment.

[0054] Step 2: Use the temperature, humidity and load intensity data of the equipment as the training set, and the corresponding equipment maintenance and repair demand data as the label. Build an equipment maintenance and repair demand prediction model based on deep learning algorithm, train the model, and obtain a trained equipment maintenance and repair demand difficulty prediction model.

[0055] Step 3: Use sensors to collect data on ambient temperature, humidity and load intensity of the equipment, and input them into the equipment maintenance and repair demand difficulty data prediction model to obtain the predicted equipment maintenance and repair demand data output by the model. Use the collected electrical variable data and the internal working temperature data of the equipment to correct the predicted equipment maintenance and repair demand difficulty data to obtain the comprehensive equipment maintenance and repair demand index.

[0056] Step 4: Based on the comprehensive equipment maintenance and repair demand index obtained above, the equipment maintenance and repair interval is calculated using a dynamic adjustment formula. According to the calculated maintenance and repair interval, the system will provide an adjustment time for maintenance and repair. When the comprehensive maintenance and repair demand index is higher than the preset maintenance and repair demand threshold, the system will increase the maintenance difficulty indicator.

[0057] By installing multiple sensors, more complete data can be collected. It can be seen that these parameters directly affect the operation of the equipment. The resulting index, obtained by inputting these related and influential data into the model, can accurately reflect the maintenance and repair needs of the equipment.

[0058] Ambient temperature and humidity sensors are installed in different locations in the environment where the equipment is located. Load sensors, current sensors, and temperature sensors in the electrical control system are installed inside the equipment to collect corresponding data. Multiple sets of ambient temperature and humidity sensors can be installed and the average value is used as the data to be recorded. Ambient temperature and humidity sensors should be kept away from the equipment to avoid the temperature and moisture generated by the equipment itself affecting the accuracy of the data.

[0059] Equipment maintenance and repair requirements data are obtained based on historical equipment maintenance and repair data;

[0060] The load intensity data of the equipment is divided into the initial operation stage, the normal operation stage, and the tail operation stage.

[0061] The data output from each sensor is in a unified format, including timestamp, sensor type, and reading value, and the data will be securely backed up to a medium.

[0062] The data from each sensor is preprocessed to ensure data quality. The specific steps are as follows:

[0063] Remove outliers: Calculate the standard deviation of the data and set 3 times the standard deviation as a threshold. Data points that exceed this range are considered outliers.

[0064] The formula used to calculate the standard deviation of the data is:

[0065]

[0066] in, This is the standard deviation, and N is the total number of data points. It is the first Data points, It is the average value of the data;

[0067] The formula used to calculate the average of the data is:

[0068]

[0069] in, N is the average of the data, and N is the total number of data points. It is the first One data point;

[0070] Data interpolation: For time series data, linear interpolation is used to fill in missing values: assuming that the data changes linearly between two known points, missing values ​​are estimated by calculating the linear relationship between these two points.

[0071] The process of establishing a predictive model for equipment maintenance and repair needs based on deep learning algorithms specifically includes:

[0072] A prediction model for equipment maintenance and repair needs is established based on deep learning algorithms. The model consists of an input layer, a hidden layer, an activation layer, and an output layer. The temperature, humidity, and load intensity data of the equipment are used as the training set, and the corresponding equipment maintenance and repair needs data are used as labels to train the model, resulting in a well-trained prediction model for equipment maintenance and repair needs.

[0073] Data processing based on deep learning algorithms can better integrate complex data, providing more complete and effective reminders of maintenance difficulty and maintenance intervals. This allows users to intuitively and proactively prepare for maintenance, thus avoiding delays in equipment maintenance that could lead to power generation interruptions. In addition, the collection of sensor data can avoid the influence of human factors and improve accuracy.

[0074] The predicted equipment maintenance and repair demand data L is corrected using the collected electrical variable data and the equipment's internal operating temperature data, resulting in the following formula for the comprehensive equipment maintenance and repair demand index:

[0075]

[0076] Where X is the comprehensive equipment maintenance and repair demand index, S is electrical variable data, L is the equipment internal operating temperature data, and H is the predicted equipment maintenance and repair difficulty data. It is the preset scaling factor for electrical variable data. It is a preset proportional coefficient for the internal operating temperature data of the equipment. It is a preset proportional coefficient for the predicted difficulty data of equipment maintenance and repair needs. , , Both are greater than zero, and > > , It is a constant correction exponent.

[0077] In this technical solution, electrical variable data and equipment internal operating temperature data are the direct factors affecting equipment maintenance. Drastic fluctuations in electrical variables directly affect equipment operation. In addition, under different circumstances, excessively high or low equipment temperature is also one of the direct influencing factors. Under normal circumstances, the probability of maintenance caused by electrical variables affecting equipment operation is 30-50%, while the probability of maintenance caused by internal operating temperature is 25-45%. The predicted equipment maintenance difficulty data usually accounts for 10-30% of the weight.

[0078] The maintenance and repair demand index for comprehensive equipment is used to calculate the maintenance and repair interval using a dynamic adjustment formula, which is as follows:

[0079]

[0080] in, It is the calculated maintenance and repair interval. This is the initial maintenance interval, in days, where X is the comprehensive equipment maintenance demand index. It is a preset adjustment coefficient, and Greater than zero.

[0081] Based on the calculated maintenance intervals, the system will automatically adjust the difficulty of equipment maintenance requirements, specifically including:

[0082] When the comprehensive equipment maintenance and repair demand index X is higher than the preset water demand threshold That is, X > At this time, the system will increase the difficulty of equipment maintenance and repair needs. The formula for calculating the maintenance and repair interval is as follows:

[0083]

[0084] in, The difficulty of equipment maintenance and repair needs X represents the basic maintenance difficulty, while X represents the comprehensive equipment maintenance and repair demand index. This is the preset proportional coefficient for the comprehensive equipment maintenance and repair demand index X. Greater than zero.

[0085] The value of X is usually between 0.7 and 0.85, and there is a positive correlation between X and B.

[0086] In the formula , , The specific values ​​are generally determined by those skilled in the art based on the actual situation. Multiple sets of sample data are collected by those skilled in the art, and a corresponding preset proportion coefficient is set for each set of sample data. The preset proportion coefficient and the collected sample data are substituted into the formula. Through repeated experiments and parameter adjustments, the accuracy of the model output and the rationality of the results are observed. These factor coefficients are gradually adjusted, and the performance and effect of the model under different parameter settings are compared to find the optimal coefficient combination. The calculated factor coefficients are then screened and averaged to obtain the final value. , , The value of .

[0087] In addition, the size of the preset factor coefficient is a specific value obtained by quantifying each parameter. In order to facilitate subsequent comparison, the size of the coefficient depends on the amount of sample data and the preset ratio coefficient initially set by those skilled in the art for each set of sample data. It is not unique, as long as it does not affect the ratio relationship between the parameter and the quantified value.

[0088] The load sensor and current sensor of the electrical control system are installed on the main circuit of the equipment to monitor the working status of the equipment. The load sensor and current sensor of the electrical control system are connected in parallel in the circuit, do not interfere with each other, and are used to collect the equipment operation information at the same time.

[0089] For the purposes of this invention, the above formula is a dimensionless calculation. The formula is a formula obtained by software simulation based on a large amount of collected data to obtain the most recent real situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0090] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for power plant production management, characterized in that, The specific steps include: Step 1: Collect the ambient temperature of the power equipment using an ambient temperature sensor; collect the ambient humidity of the power equipment using an ambient humidity sensor; collect the load intensity of the equipment using an electrical control system load sensor; measure the electrical variables in the equipment's operating status using a current sensor; collect the internal operating temperature of the equipment and data on the difficulty of equipment maintenance and repair using a temperature sensor. Step 2: Use the temperature, humidity and load intensity data of the equipment as the training set, and the corresponding equipment maintenance and repair demand data as the label. Build an equipment maintenance and repair demand prediction model based on deep learning algorithm, train the model, and obtain a trained equipment maintenance and repair demand difficulty prediction model. Step 3: Use sensors to collect data on ambient temperature, humidity and load intensity of the equipment, and input them into the equipment maintenance and repair demand difficulty data prediction model to obtain the predicted equipment maintenance and repair demand data output by the model. Use the collected electrical variable data and the internal working temperature data of the equipment to correct the predicted equipment maintenance and repair demand difficulty data to obtain the comprehensive equipment maintenance and repair demand index. Step 4: Based on the comprehensive equipment maintenance and repair demand index obtained above, the equipment maintenance and repair interval is calculated using a dynamic adjustment formula. According to the calculated maintenance and repair interval, the system will provide an adjustment time for maintenance and repair. When the comprehensive maintenance and repair demand index is higher than the preset maintenance and repair demand threshold, the system will increase the maintenance difficulty indicator. The data from each sensor is preprocessed to ensure data quality. The specific steps are as follows: Remove outliers: Calculate the standard deviation of the data and set 3 times the standard deviation as a threshold. Data points that exceed this range are considered outliers. The formula used to calculate the standard deviation of the data is: ; in, It is the standard deviation. It is the total number of data points. It is the first Data points, It is the average value of the data; The formula used to calculate the average of the data is: ; in, It is the average of the data. It is the total number of data points. It is the first One data point; Data interpolation: For time series data, linear interpolation is used to fill in missing values: assuming that the data changes linearly between two known points, missing values ​​are estimated by calculating the linear relationship between these two points; The process of establishing a predictive model for equipment maintenance and repair needs based on deep learning algorithms specifically includes: A deep learning algorithm-based model for predicting equipment maintenance and repair needs difficulty is established. This model consists of an input layer, hidden layer, activation layer, and output layer. Temperature, humidity, and load intensity data of the equipment are used as the training set, and corresponding equipment maintenance and repair needs data are used as labels to train the model, resulting in a well-trained model for predicting equipment maintenance and repair needs difficulty. The predicted equipment maintenance and repair needs data L is then corrected using collected electrical variable data and internal operating temperature data of the equipment, yielding the formula for the comprehensive equipment maintenance and repair needs index, as follows: ; Where X is the comprehensive equipment maintenance and repair demand index, S is electrical variable data, L is the equipment internal operating temperature data, and H is the predicted equipment maintenance and repair difficulty data. It is the preset scaling factor for electrical variable data. It is a preset proportional coefficient for the internal operating temperature data of the equipment. It is a preset proportional coefficient for the predicted difficulty data of equipment maintenance and repair needs. , , Both are greater than zero, and > > , The constant correction exponent; The comprehensive equipment maintenance and repair demand index is calculated using a dynamic adjustment formula to determine the maintenance and repair interval. The formula is as follows: ; in, It is the calculated maintenance and repair interval. This is the initial maintenance interval, in days. It is a comprehensive equipment maintenance and repair demand index. It is a preset adjustment coefficient, and Greater than zero.

2. The method for power plant production management according to claim 1, characterized in that, The ambient temperature sensor and ambient humidity sensor are installed at different locations in the environment where the equipment is located. The load sensor, current sensor and temperature sensor of the electrical control system are installed inside the equipment to collect corresponding data. Multiple sets of ambient temperature sensors and ambient humidity sensors are installed and the average value is used as the data to be recorded. The ambient temperature sensor and ambient humidity sensor should be kept away from the equipment to avoid the temperature and moisture generated by the equipment itself from affecting the accuracy of the data. The equipment maintenance and repair requirements data are obtained based on historical equipment maintenance and repair data. The load intensity data of the equipment is divided into the initial operation stage, the normal operation stage, and the tail operation stage.

3. The method for power plant production management according to claim 2, characterized in that, The data output from each sensor is in a unified format, including timestamp, sensor type, and reading value, and the data will be securely backed up to a medium.

4. The method for power plant production management according to claim 1, characterized in that: Based on the calculated maintenance intervals, the system will automatically adjust the difficulty of equipment maintenance requirements, specifically including: When the comprehensive equipment maintenance and repair demand index Higher than the preset water requirement threshold ,Right now > At this time, the system will increase the difficulty of equipment maintenance and repair needs. The formula for calculating the maintenance and repair interval is as follows: ; in, The difficulty lies in the equipment maintenance and repair requirements. It is the difficulty of basic maintenance. This is a comprehensive equipment maintenance and repair demand index. The comprehensive equipment maintenance and repair demand index The preset proportional coefficient, Greater than zero.

Images