An artificial intelligence-based industrial enterprise energy management method and system

Abstract

The application relates to the technical field of energy management, and discloses an industrial enterprise energy management method and system based on artificial intelligence, which comprises a multidimensional acquisition module and an intelligent evaluation module. The industrial enterprise energy management method and system based on artificial intelligence acquire the operation data of all industrial equipment and the consumption records of different kinds of energy through the multidimensional acquisition module, and classify and form a data set; the intelligent evaluation module analyzes the power ratio, production efficiency and maintenance frequency of each industrial equipment according to the operation state of the industrial equipment, generates a decay coefficient, discovers aging damage risks in time, sets a fixed monitoring period, analyzes the peak-valley difference in the consumption process of each kind of energy, generates an adaptation coefficient, quantifies the adaptation degree of each kind of energy, and more accurately evaluates the energy consumption level; the intelligent evaluation module judges the performance decay degree, energy supply state and adaptation degree of the industrial equipment, and outputs corresponding early warning signals, so that the energy supply is dynamically adjusted and is more stable.

Description

Technical Field

[0001] This invention relates to the field of energy management technology, specifically to an energy management method and system for industrial enterprises based on artificial intelligence. Background Technology

[0002] With the global energy crisis and increasingly stringent environmental requirements, energy management has become particularly important in the industrial sector. Industrial enterprise energy management refers to optimizing energy use efficiency, reducing energy consumption, and minimizing negative environmental impacts through scientific planning, effective implementation, and strict supervision. The types of energy required for industrial production include fossil fuels, hydropower, electricity, heat, liquid fuels, renewable energy, nuclear energy, industrial waste heat, hydrogen energy, and compressed air energy. The use of each type of energy depends on production processes, equipment requirements, and energy supply conditions. The core of energy use management is ensuring the rational allocation and use of energy, reducing energy waste through the rational scheduling of energy demand at each stage of the production process. It not only helps enterprises reduce production costs but also enhances their social responsibility and strengthens their competitiveness.

[0003] Currently, traditional industrial enterprises' energy management methods neglect the impact of equipment aging, wear and tear or failure on energy consumption. Industrial equipment lacking effective maintenance will not only experience performance decline but also lead to energy waste and even higher maintenance costs. In addition, when energy demand fluctuates, there is a lack of a unified assessment mechanism, making it impossible to replace the supply of energy in a timely manner, and making it difficult to ensure cost control and supply security of energy consumption. Summary of the Invention

[0004] (a) Technical problems to be solved

[0005] To address the shortcomings of existing technologies, this invention provides an artificial intelligence-based energy management method and system for industrial enterprises. It has advantages such as multi-dimensional assessment of equipment energy consumption levels and dynamic adjustment of energy supply for greater stability. It solves the problem that traditional industrial enterprise energy management methods neglect the impact of equipment aging, wear, or failure on energy consumption and make it difficult to ensure a stable supply.

[0006] (II) Technical Solution

[0007] To achieve the above objectives, the present invention provides the following technical solution: an energy management method for industrial enterprises based on artificial intelligence, comprising the following steps:

[0008] Step 1: Acquire the operating data of all industrial equipment and the consumption records of different types of energy through the central control console and sensing devices, and classify them into equipment datasets and real-time datasets;

[0009] Step 2: Based on the equipment dataset, analyze the power ratio, production efficiency, and maintenance frequency of each industrial equipment according to its operating status, and generate the corresponding degradation coefficient Stx;

[0010] Step 3: Set a fixed monitoring period Q, and then combine it with the real-time dataset to analyze the peak-valley difference Fhc in each energy consumption process and generate the corresponding adaptation coefficient Spx.

[0011] Step 4: Set fixed ranges for the degradation threshold STY, consumption threshold XLY, cost threshold JGY, correlation threshold GSY, recovery threshold HSY, and adaptation threshold SPY. Then, combine these with the degradation coefficient Stx, peak-to-valley difference Fhc, and adaptation coefficient Spx to determine the degree of performance degradation of industrial equipment, energy supply status, and the degree of energy supply adaptation, and output corresponding early warning signals.

[0012] Preferably, in step one, the expression for the device dataset is {S1}. d S2 d S3 d ... Sn d}, S1 d To Sn d These are the operating data for the first to the nth industrial equipment, respectively. The operating data includes operating status, no-load power, load power, output, installation time, and number of maintenance operations. d represents the specific time when the production data of each equipment was acquired.

[0013] Preferably, in step one, the expression for the real-time dataset is {N1}. m N2 m N3 m ... Ny m}, N1 m To Ny m These are the consumption records for the first to the yth energy sources, respectively. Each consumption record includes the consumption amount, price cost, number of associated devices, and recyclability rate. m represents the specific time when the consumption records for each energy source are obtained.

[0014] Preferably, in step two, the calculation process for the degradation coefficient Stx is as follows:

[0015] Extract the operating data of the i-th industrial equipment from the equipment dataset, and label the no-load power of the i-th industrial equipment as KP. i Let the load power of the i-th industrial device be denoted as FP. i Let CL be the output of the i-th industrial equipment. i The installation time point of the i-th industrial equipment is marked as AZ. i The number of repairs for the i-th industrial equipment is marked as WX. i ;

[0016] If the i-th industrial equipment is in an unloaded state.

[0017]

[0018] In the formula, D represents the current time point, and D-AZ i This represents the usage time of the i-th industrial device. The no-load ratio is the ratio of no-load power to usage time. α1 represents the evaluation weight for the no-load ratio. The ratio of the number of repairs to the usage time is called the repair frequency. α2 represents the evaluation weight for the repair frequency, and α1 + α2 = 1. This indicates that the degradation coefficient of the i-th industrial equipment is calculated according to the weights α1 and α2.

[0019] If the i-th industrial device is in a loaded state...

[0020]

[0021] In the formula, The load ratio is the ratio of load power to usage time. β1 represents the evaluation weight of the load ratio. β1 represents the ratio of load power to output, which is the production efficiency. β2 represents the evaluation weight for production efficiency, and β3 represents the evaluation weight for maintenance frequency. β1 + β2 + β3 = 1. This indicates that the degradation coefficient of the i-th industrial equipment is calculated according to the weights β1, β2, and β3.

[0022] Preferably, in step three, the peak-to-valley difference Fhc calculation process is as follows:

[0023] Based on the real-time dataset, the consumption records of the k-th energy source are statistically analyzed within the monitoring period Q, and the consumption of the k-th energy source is labeled as {xl1, xl2, xl3, ..., xl...} a}, xl1 to xl a Let {jg1, jg2, jg3, ..., ig} represent the consumption from the first time point to the a-th time point, and let {jg1, jg2, jg3, ..., ig} be the price cost of the k-th energy source. a}, jg1 to jg a Let gs1, gs2, gs3, ..., gsn be the price and cost from the first time point to the a-th time point, and let gs2, gs3, ..., gsn be the associated equipment for the k-th energy source. a}, gs1 to gs a Let be the number of associated devices from the first time point to the a-th time point, and let the recyclability of the k-th energy source be denoted as {hs1, hs2, hs3, ..., hs...}.a}, hs1 to hs a These represent the recoverability rates from the first time point to the a-th time point, respectively.

[0024]

[0025] In the formula, maxxl and minxl are the maximum and minimum values ​​of the consumption of the k-th energy source within the monitoring period Q, respectively; maxxl-minxl represents the range of the consumption of the k-th energy source; maxjg and minjg are the highest and lowest values ​​of the price cost of the k-th energy source within the monitoring period Q, respectively; maxjg-minjg represents the range of the price cost of the k-th energy source; maxgs and mings are the maximum and minimum values ​​of the number of devices associated with the k-th energy source within the monitoring period Q, respectively; maxgs-mings represents the range of the number of devices associated with the k-th energy source; maxhs and minhs are the highest and lowest values ​​of the recyclability rate of the k-th energy source within the monitoring period Q, respectively; maxhs-minhs represents the range of the recyclability rate of the k-th energy source.

[0026] Preferably, in step three, the adaptation coefficient Spx is calculated as follows:

[0027]

[0028] In the formula, This represents the average recyclability rate of the k-th energy source within the monitoring period Q. ω1 represents the ratio of the consumption of the k-th energy type to its price cost, which is the price cost required per unit of consumption. ω1 represents the evaluation weight of the price cost required per unit of consumption. Let ω1 represent the ratio of the k-th energy consumption to the number of associated devices, which is the average energy consumption level of each associated device. ω2 represents the evaluation weight for the average energy consumption level, and ω3 represents the evaluation weight for the average energy recoverability rate. ω1 + ω2 + ω3 = 1. This indicates that the adaptation coefficient of the k-th energy source is calculated according to the weights of ω1, ω2, and ω3.

[0029] Preferably, in step four, when the degradation coefficient Stx exceeds the degradation threshold STY, it indicates that the performance degradation of the industrial equipment is severe, and a warning signal is generated.

[0030] Preferably, in step four, when the range of energy consumption exceeds the consumption threshold XLY, the range of energy price cost exceeds the cost threshold JGY, the range of the number of energy-related devices exceeds the association threshold GSY, or the range of energy recoverability is lower than the recovery threshold HSY, it indicates that the energy supply status is unstable and a second warning signal is generated.

[0031] Preferably, in step four, when the adaptation coefficient Spx is lower than the adaptation threshold SPY, it indicates that the adaptation degree of the energy supply is low, and a warning signal three is generated. The processing priority of warning signal one is higher than that of warning signal two, and the processing priority of warning signal two is higher than that of warning signal three.

[0032] An artificial intelligence-based energy management system for industrial enterprises includes a multi-dimensional data acquisition module and an intelligent assessment module;

[0033] The multi-dimensional acquisition module consists of an operation data unit and an energy data unit. The operation data unit collects equipment datasets via a network connection to a central control console. The equipment datasets include the operation data of all industrial equipment. The energy data unit collects real-time datasets via a network connection to sensor devices. The real-time datasets include consumption records of different types of energy.

[0034] The intelligent evaluation module consists of a performance evaluation unit, an energy consumption evaluation unit, and an early warning management unit. The performance evaluation unit analyzes the power ratio, production efficiency, and maintenance frequency of each industrial device based on the equipment dataset and the operating status of the industrial equipment, generating a corresponding degradation coefficient Stx. The energy consumption evaluation unit has a fixed monitoring period Q and, combined with real-time datasets, analyzes the peak-to-valley difference Fhc in each energy consumption process, generating a corresponding adaptation coefficient Spx. The early warning management unit has fixed-range degradation thresholds STY, consumption threshold XLY, cost threshold JGY, correlation threshold GSY, recovery threshold HSY, and adaptation threshold SPY. Combining the degradation coefficient Stx, peak-to-valley difference Fhc, and adaptation coefficient Spx, it determines the degree of performance degradation, energy supply status, and energy supply adaptation of the industrial equipment, and outputs corresponding early warning signals.

[0035] Compared with existing technologies, this invention provides an artificial intelligence-based energy management method and system for industrial enterprises, which has the following beneficial effects:

[0036] 1. This invention uses a multi-dimensional acquisition module network to connect a central control console and sensing devices to acquire operating data of all industrial equipment and consumption records of different types of energy. This data is then categorized into equipment datasets and real-time datasets. The intelligent evaluation module, based on the equipment datasets and the operating status of each industrial device, analyzes its power ratio, production efficiency, and maintenance frequency, generating a corresponding degradation coefficient Stx. This allows for timely detection of equipment aging or damage risks, identifying industrial equipment requiring priority repair or optimization, and effectively preventing production interruptions caused by equipment failures. The intelligent evaluation module sets a fixed monitoring period Q and, combined with the real-time dataset, analyzes the peak-to-valley difference Fhc during the consumption process of each energy source, generating a corresponding adaptation coefficient Spx. This establishes a unified evaluation mechanism, quantifying the adaptability of each energy source, which helps improve the matching degree between energy supply and equipment, resulting in more accurate multi-dimensional assessments of energy consumption levels.

[0037] 2. This invention uses an intelligent evaluation module to set fixed ranges for the following thresholds: STY (degradation threshold), XLY (consumption threshold), JGY (cost threshold), GSY (association threshold), HSY (recovery threshold), and SPY (adaptation threshold). Combined with the degradation coefficient Stx, peak-to-valley difference Fhc, and adaptation coefficient Spx, it determines the degree of performance degradation of industrial equipment, the energy supply status, and the degree of energy supply adaptation, and outputs corresponding early warning signals. Early warning signal one has a higher processing priority than early warning signal two, and early warning signal two has a higher processing priority than early warning signal three, ensuring the efficiency and reliability of energy supply, effectively avoiding energy waste, and dynamically adjusting energy supply for greater stability. Attached Figure Description

[0038] Figure 1 This is a diagram illustrating the steps of the method of the present invention;

[0039] Figure 2 This is a schematic diagram of the system flow of the present invention. Detailed Implementation

[0040] 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.

[0041] Traditional industrial energy management methods neglect the impact of equipment aging, wear, or malfunctions on energy consumption. Industrial equipment lacking effective maintenance not only experiences performance degradation but also leads to energy waste and even higher maintenance costs. Furthermore, the lack of a unified assessment mechanism during energy demand fluctuations hinders timely energy supply replacement, making it difficult to guarantee cost control and supply security. Therefore, this paper presents an artificial intelligence-based industrial energy management method and system. Please refer to [link / reference]. Figure 1 An artificial intelligence-based energy management method for industrial enterprises includes the following steps:

[0042] Step 1: Acquire the operating data of all industrial equipment and the consumption records of different types of energy through the central control console and sensing devices, and classify them into equipment datasets and real-time datasets;

[0043] The expression for the device dataset is {S1} d S2 d S3 d ... Sn d}, S1 d To Sn d These are the operating data for the first to the nth industrial equipment, including operating status, no-load power, load power, output, installation time, and number of maintenance operations. d represents the specific time when the production data for each piece of equipment was acquired. Many industrial equipment continue to operate under low load, idling, or during unnecessary processes, resulting in significant differences between no-load power and load power.

[0044] The expression for the real-time dataset is {N1} m N2 m N3 m ... Ny m}, N1 m To Ny m These are the consumption records for the first to the yth energy sources, respectively. The consumption records include the consumption amount, price cost, number of associated equipment, and recycling rate. m represents the specific time when the consumption records for each energy source are acquired. Comprehensive collection of energy consumption records allows for more refined energy management in the future.

[0045] Step 2: Based on the equipment dataset, analyze the power ratio, production efficiency, and maintenance frequency of each industrial piece of equipment according to its operating status, and generate the corresponding degradation coefficient Stx. The calculation process is as follows:

[0046] Extract the operating data of the i-th industrial equipment from the equipment dataset, and label the no-load power of the i-th industrial equipment as KP. i Let the load power of the i-th industrial device be denoted as FP. i Let CL be the output of the i-th industrial equipment.i The installation time point of the i-th industrial equipment is marked as AZ. i The number of repairs for the i-th industrial equipment is marked as WX. i ;

[0047] If the i-th industrial equipment is in an unloaded state.

[0048]

[0049] In the formula, D represents the current time point, and D-AZ i This represents the usage time of the i-th industrial device. The no-load ratio is the ratio of no-load power to usage time. α1 represents the evaluation weight for the no-load ratio. The ratio of the number of repairs to the usage time is called the repair frequency. α2 represents the evaluation weight for the repair frequency, and α1 + α2 = 1. This indicates that the degradation coefficient of the i-th industrial equipment is calculated according to the weights α1 and α2.

[0050] If the i-th industrial device is in a loaded state...

[0051]

[0052] In the formula, The load ratio is the ratio of load power to usage time. β1 represents the evaluation weight of the load ratio. β1 represents the ratio of load power to output, which is the production efficiency. β2 represents the evaluation weight for production efficiency, and β3 represents the evaluation weight for maintenance frequency. β1 + β2 + β3 = 1. This means that the degradation coefficient of the i-th industrial equipment is calculated according to the weights of β1, β2 and β3, so as to promptly detect the risk of equipment aging or damage, identify the industrial equipment that needs to be repaired or optimized in a timely manner, and effectively avoid production interruptions caused by equipment failure.

[0053] Step 3: Set a fixed monitoring period Q, and then combine it with the real-time dataset to analyze the peak-valley difference Fhc in each energy consumption process and generate the corresponding adaptation coefficient Spx.

[0054] The calculation process for peak-to-valley difference Fhc is as follows:

[0055] Based on the real-time dataset, the consumption records of the k-th energy source are statistically analyzed within the monitoring period Q, and the consumption of the k-th energy source is labeled as {xl1, xl2, xl3, ..., xl...} a}, xl1 to xl aLet {jg1, jg2, jg3, ..., jg} represent the consumption from the first time point to the a-th time point, and let {jg1, jg2, jg3, ..., jg} be the price cost of the k-th energy source. a}, jg1 to jg a Let gs1, gs2, gs3, ..., gsn be the price and cost from the first time point to the a-th time point, and let gs2, gs3, ..., gsn be the associated equipment for the k-th energy source. a}, gs1 to gs a Let be the number of associated devices from the first time point to the a-th time point, and let the recyclability of the k-th energy source be denoted as {hs1, hs2, hs3, ..., hs...}. a}, hs1 to hs a These represent the recoverability rates from the first time point to the a-th time point, respectively.

[0056]

[0057] In the formula, maxxl and minxl represent the maximum and minimum values ​​of the consumption of the k-th energy type within the monitoring period Q, respectively; maxxl - minxl represents the range of the consumption of the k-th energy type; maxjg and minjg represent the highest and lowest values ​​of the price cost of the k-th energy type within the monitoring period Q, respectively; maxjg - minjg represents the range of the price cost of the k-th energy type; maxgs and mings represent the maximum and minimum values ​​of the number of devices associated with the k-th energy type within the monitoring period Q, respectively; maxgs - mings represents the range of the number of devices associated with the k-th energy type; maxhs and minhs represent the highest and lowest values ​​of the recyclability rate of the k-th energy type within the monitoring period Q, respectively; maxhs - minhs represents the range of the recyclability rate of the k-th energy type. This formula intuitively reflects the peaks and troughs that occur during the consumption of different types of energy, records the most unstable time points in energy consumption, and provides a basis for subsequent optimization of energy procurement, supply, and use.

[0058] The adaptation factor SpX is calculated as follows:

[0059]

[0060] In the formula, This represents the average recyclability rate of the k-th energy source within the monitoring period Q. ω1 represents the ratio of the consumption of the k-th energy type to its price cost, which is the price cost required per unit of consumption. ω1 represents the evaluation weight of the price cost required per unit of consumption. Let ω1 represent the ratio of the k-th energy consumption to the number of associated devices, which is the average energy consumption level of each associated device. ω2 represents the evaluation weight for the average energy consumption level, and ω3 represents the evaluation weight for the average energy recoverability rate. ω1 + ω2 + ω3 = 1. This means that the adaptation coefficient of the k-th energy source is calculated according to the weights of ω1, ω2, and ω3, and a unified evaluation mechanism is established to quantify the adaptation degree of each energy source, which helps to improve the matching degree between energy supply and equipment.

[0061] Step 4: Set fixed ranges for the following thresholds: STY (degradation threshold), XLY (consumption threshold), JGY (cost threshold), GSY (correlation threshold), HSY (recovery threshold), and SPY (adaptation threshold). Combine these with the degradation coefficient Stx, peak-to-valley difference Fhc, and adaptation coefficient Spx to determine the degree of performance degradation of the industrial equipment, its energy supply status, and the degree of energy supply adaptation. When the degradation coefficient Stx exceeds the degradation threshold STY, it indicates a severe performance degradation of the industrial equipment, generating warning signal one. In the peak-to-valley difference Fhc, the range of energy consumption exceeds the consumption threshold XLY, and the range of energy price cost exceeds the cost threshold. When the range of JGY and the number of energy-related devices exceeds the correlation threshold GSY, or the range of energy recoverability is lower than the recovery threshold HSY, it indicates that the energy supply is at risk of instability, generating warning signal two. When the adaptation coefficient Spx is lower than the adaptation threshold SPY, it indicates that the energy supply is poorly adapted, generating warning signal three. Adjustments are made to energy scheduling during peak periods or alternative energy sources are used. Warning signal one has a higher processing priority than warning signal two, and warning signal two has a higher processing priority than warning signal three. Dynamically adjust energy usage strategies to ensure the efficiency and reliability of energy supply and effectively avoid energy waste.

[0062] Please see Figure 2 An artificial intelligence-based energy management system for industrial enterprises, comprising a multi-dimensional data acquisition module and an intelligent assessment module;

[0063] The multidimensional acquisition module consists of an operational data unit and an energy data unit. The operational data unit collects equipment datasets via a network connection to the central control console. The equipment datasets include operational data from all industrial equipment. The energy data unit collects real-time datasets via a network connection to sensor devices. The real-time datasets include consumption records of different types of energy.

[0064] The intelligent assessment module consists of a performance assessment unit, an energy consumption assessment unit, and an early warning management unit. The performance assessment unit analyzes the power ratio, production efficiency, and maintenance frequency of each industrial device based on the equipment dataset and the operating status of the industrial equipment, generating a corresponding degradation coefficient Stx. The energy consumption assessment unit has a fixed monitoring period Q and, combined with real-time datasets, analyzes the peak-to-valley difference Fhc in each energy consumption process, generating a corresponding adaptation coefficient Spx. This multi-dimensional assessment of energy consumption levels is more accurate. The multi-dimensional assessment of equipment energy consumption levels and early warning management unit have fixed-range degradation thresholds STY, consumption threshold XLY, cost threshold JGY, correlation threshold GSY, recovery threshold HSY, and adaptation threshold SPY. Combining the degradation coefficient Stx, peak-to-valley difference Fhc, and adaptation coefficient Spx, it determines the degree of performance degradation, energy supply status, and energy supply adaptation of the industrial equipment, outputting corresponding early warning signals and dynamically adjusting the energy supply for greater stability.

[0065] Example 1: In this experiment, an industrial device under load was selected as the experimental object. Monitoring showed that the device had been in production for 80 months, with 30 maintenance cycles. Under one-hour load, the device's power output was 50kW, and its production capacity was 150 units. The degradation coefficient Stx of this industrial device was calculated using the following formula:

[0066]

[0067] In the formula, The load ratio is the ratio of load power to usage time. 0.4 indicates the weighting of the load ratio in the evaluation. The ratio of load power to output is the production efficiency. 0.4 represents the evaluation weight for production efficiency, and 0.2 represents the evaluation weight for maintenance frequency. β1+β2+β3=1. Based on the weights of β1, β2 and β3, the degradation coefficient of the industrial equipment is calculated to be 0.46.

[0068] Example 2: In this experiment, coal energy was selected as the experimental object, and the monitoring period was set to 30 days. Monitoring showed that the maximum daily coal consumption was 500 tons, and the minimum was 200 tons. The highest daily coal price was 1200 yuan / ton, and the lowest was 800 yuan / ton. The maximum number of coal-related devices was 50, and the minimum was 30. The highest coal recovery rate was 85%, and the lowest was 60%. The formula for calculating the coal peak-to-valley difference Fhc is as follows:

[0069]

[0070] In the formula, 300 tons represents the range of coal energy consumption, 400 yuan / ton represents the range of coal energy price cost, 20 units represents the range of the number of coal energy-related equipment, and 25% represents the range of coal energy recoverability.

[0071] 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. An energy management method for industrial enterprises based on artificial intelligence, characterized in that, Includes the following steps: Step 1: Acquire the operating data of all industrial equipment and the consumption records of different types of energy through the central control console and sensing devices, and classify them into equipment datasets and real-time datasets; Step 2: Based on the equipment dataset, analyze the power ratio, production efficiency, and maintenance frequency of each industrial equipment according to its operating status, and generate the corresponding degradation coefficient Stx; The calculation process for the degradation coefficient Stx is as follows: Extract the operating data of the i-th industrial equipment from the equipment dataset, and label the no-load power of the i-th industrial equipment as KP. i Let the load power of the i-th industrial device be denoted as FP. i Let CL be the output of the i-th industrial equipment. i The installation time point of the i-th industrial equipment is marked as AZ. i The number of repairs for the i-th industrial equipment is marked as WX. i ; If the i-th industrial equipment is in an unloaded state. In the formula, D represents the current time point, and D-AZ i This represents the usage time of the i-th industrial device. The no-load ratio is the ratio of no-load power to usage time. α1 represents the evaluation weight for the no-load ratio. The ratio of the number of repairs to the usage time is called the repair frequency. α2 represents the evaluation weight for the repair frequency, and α1 + α2 = 1. This indicates that the degradation coefficient of the i-th industrial equipment is calculated according to the weights α1 and α2. If the i-th industrial device is in a loaded state... In the formula, The load ratio is the ratio of load power to usage time. β1 represents the evaluation weight of the load ratio. β1 represents the ratio of load power to output, which is the production efficiency. β2 represents the evaluation weight for production efficiency, and β3 represents the evaluation weight for maintenance frequency. β1 + β2 + β3 = 1. This indicates that the degradation coefficient of the i-th industrial equipment is calculated according to the weights β1, β2, and β3. Step 3: Set a fixed monitoring period Q, and then combine it with the real-time dataset to analyze the peak-valley difference Fhc in each energy consumption process and generate the corresponding adaptation coefficient Spx. The calculation process for peak-to-valley difference Fhc is as follows: Based on the real-time dataset, the consumption records of the k-th energy source are statistically analyzed within the monitoring period Q, and the consumption of the k-th energy source is labeled as {xl1, xl2, xl3, ..., xl...} a }, xl1 to xl a Let {jg1, jg2, jg3, ..., jg} represent the consumption from the first time point to the a-th time point, and let {jg1, jg2, jg3, ..., jg} be the price cost of the k-th energy source. a }, jg1 to jg a Let gs1, gs2, gs3, ..., gsn be the price and cost from the first time point to the a-th time point, and let gs2, gs3, ..., gsn be the associated equipment for the k-th energy source. a }, gs1 to gs a Let be the number of associated devices from the first time point to the a-th time point, and let the recyclability of the k-th energy source be denoted as {hs1, hs2, hs3, ..., hs...}. a }, hs1 to hs a These represent the recoverability rates from the first time point to the a-th time point, respectively. In the formula, maxxl and minxl are the maximum and minimum values ​​of the consumption of the k-th energy source within the monitoring period Q, respectively; maxxl-minxl represents the range of the consumption of the k-th energy source; maxjg and minjg are the highest and lowest values ​​of the price cost of the k-th energy source within the monitoring period Q, respectively; maxjg-minjg represents the range of the price cost of the k-th energy source; maxgs and mings are the maximum and minimum values ​​of the number of devices associated with the k-th energy source within the monitoring period Q, respectively; maxgs-mings represents the range of the number of devices associated with the k-th energy source; maxhs and minhs are the highest and lowest values ​​of the recyclability rate of the k-th energy source within the monitoring period Q, respectively; maxhs-minhs represents the range of the recyclability rate of the k-th energy source. The adaptation factor SpX is calculated as follows: In the formula, This represents the average recoverability rate of the k-th energy source within the monitoring period Q. ω1 represents the ratio of the consumption of the k-th energy type to its price cost, which is the price cost required per unit of consumption. ω1 represents the evaluation weight of the price cost required per unit of consumption. Let ω1 represent the ratio of the k-th energy consumption to the number of associated devices, which is the average energy consumption level of each associated device. ω2 represents the evaluation weight for the average energy consumption level, and ω3 represents the evaluation weight for the average energy recoverability rate. ω1 + ω2 + ω3 = 1. This indicates that the adaptation coefficient of the k-th energy source is calculated according to the weights ω1, ω2, and ω3. Step 4: Set fixed ranges for the degradation threshold STY, consumption threshold XLY, cost threshold JGY, correlation threshold GSY, recovery threshold HSY, and adaptation threshold SPY. Then, combine these with the degradation coefficient Stx, peak-to-valley difference Fhc, and adaptation coefficient Spx to determine the degree of performance degradation of industrial equipment, energy supply status, and the degree of energy supply adaptation, and output corresponding early warning signals.

2. The energy management method for industrial enterprises based on artificial intelligence according to claim 1, characterized in that: In step one, the expression for the device dataset is {S1}. d S2 d S3 d ... Sn d }, S1 d To Sn d These are the operating data for the first to the nth industrial equipment, respectively. The operating data includes operating status, no-load power, load power, output, installation time, and number of maintenance operations. d represents the specific time when the production data of each equipment was acquired.

3. The energy management method for industrial enterprises based on artificial intelligence according to claim 2, characterized in that: In step one, the expression for the real-time dataset is {N1}. m N2 m N3 m ... Ny m }, N1 m To Ny m These are the consumption records for the first to the yth energy sources, respectively. Each consumption record includes the consumption amount, price cost, number of associated devices, and recyclability rate. m represents the specific time when the consumption records for each energy source are obtained.

4. The energy management method for industrial enterprises based on artificial intelligence according to claim 3, characterized in that: In step four, when the degradation coefficient Stx exceeds the degradation threshold STY, it indicates that the performance degradation of the industrial equipment is severe, and a warning signal is generated.

5. The energy management method for industrial enterprises based on artificial intelligence according to claim 4, characterized in that: In step four, when the range of energy consumption exceeds the consumption threshold XLY, the range of energy price cost exceeds the cost threshold JGY, the range of the number of energy-related devices exceeds the association threshold GSY, or the range of energy recoverability is lower than the recovery threshold HSY, it indicates that the energy supply status is unstable and generates warning signal two.

6. The energy management method for industrial enterprises based on artificial intelligence according to claim 5, characterized in that: In step four, when the adaptation coefficient Spx is lower than the adaptation threshold SPY, it indicates that the adaptation degree of the energy supply is low, and a warning signal three is generated. The processing priority of warning signal one is higher than that of warning signal two, and the processing priority of warning signal two is higher than that of warning signal three.

7. An artificial intelligence-based industrial enterprise energy management system, applied to the artificial intelligence-based industrial enterprise energy management method described in any one of claims 1-6, characterized in that: Includes a multi-dimensional data acquisition module and an intelligent evaluation module; The multi-dimensional acquisition module consists of an operation data unit and an energy data unit. The operation data unit collects equipment datasets via a network connection to a central control console. The equipment datasets include the operation data of all industrial equipment. The energy data unit collects real-time datasets via a network connection to sensor devices. The real-time datasets include consumption records of different types of energy. The intelligent evaluation module consists of a performance evaluation unit, an energy consumption evaluation unit, and an early warning management unit. The performance evaluation unit analyzes the power ratio, production efficiency, and maintenance frequency of each industrial device based on the equipment dataset and the operating status of the industrial equipment, generating a corresponding degradation coefficient Stx. The energy consumption evaluation unit has a fixed monitoring period Q and, combined with real-time datasets, analyzes the peak-to-valley difference Fhc in each energy consumption process, generating a corresponding adaptation coefficient Spx. The early warning management unit has fixed-range degradation thresholds STY, consumption threshold XLY, cost threshold JGY, correlation threshold GSY, recovery threshold HSY, and adaptation threshold SPY. Combining the degradation coefficient Stx, peak-to-valley difference Fhc, and adaptation coefficient Spx, it determines the degree of performance degradation, energy supply status, and energy supply adaptation of the industrial equipment, and outputs corresponding early warning signals.

Images