Oil processing energy consumption monitoring system

Abstract

The present invention relates to the field of energy consumption monitoring technology, specifically to an oil and fat processing energy consumption monitoring system, the system comprising a temperature difference input sensing module, an energy consumption fluctuation extraction module, an abnormal attribution identification module, a single-point mutation screening module, and a partition warning module. The present invention, through in-depth analysis of energy consumption data within a continuous time period, can identify specific sections with abnormal energy consumption, effectively prevent energy waste and the risk of equipment overload, and provide data support for equipment maintenance and energy conservation by analyzing the relationship between the energy consumption and working mode of equipment operation. By recording and analyzing the specific data of the abnormal sections in detail, energy management personnel can quickly locate the cause of the problem, shorten the response time, and formulate more targeted improvement measures. The monitoring of single-point mutations refines the monitoring granularity, so that the energy consumption of each key equipment is strictly monitored, reducing the chain reaction risk caused by abnormal energy consumption.

Description

Technical Field

[0001] The present invention relates to the technical field of energy consumption monitoring, in particular to an energy consumption monitoring system for oil processing. Background Art

[0002] The field of energy consumption monitoring technology involves the dynamic collection, processing, analysis, and management of various energy usage scenarios, primarily including the monitoring and evaluation of electricity, gas, steam, water, and thermal energy. This technology encompasses hardware sensor deployment, data acquisition terminals, data transmission networks, edge computing nodes, and platform-based software analysis systems, and is widely used in scenarios such as industrial manufacturing, building management, transportation, agriculture, and public facilities. Technical approaches include multi-point data collection, protocol parsing, edge fusion processing, energy consumption indicator modeling, power consumption trend forecasting, energy efficiency comparison analysis, and energy consumption anomaly warnings, enabling energy usage visualization, anomaly tracing, system optimization, and energy efficiency improvement.

[0003] The oil and fat processing energy consumption monitoring system is used to monitor, record, and analyze energy consumption data for each production unit during the oil and fat processing process. Its uses include real-time collection of energy data such as electricity, heat, and water consumption at each stage of oil and fat processing, identifying peak energy consumption, low energy efficiency, or abnormal equipment operating conditions. This system then assists management personnel in precise energy scheduling, energy efficiency assessment, and implementation of energy-saving improvement measures, thereby reducing energy consumption per unit of product and improving energy efficiency.

[0004] Traditional monitoring systems focus primarily on overall energy consumption statistics and monitoring, failing to delve into the specific energy consumption dynamics of each individual process section or piece of equipment. This results in delayed responses to energy consumption anomalies and an inability to effectively predict and control peak energy consumption, impacting overall production efficiency and cost control. For example, failure to promptly identify abnormal equipment operating conditions can lead to overloaded equipment, increasing energy consumption, shortening equipment lifespan, and increasing maintenance costs. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose an oil processing energy consumption monitoring system.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: an oil processing energy consumption monitoring system, the system comprising:

[0007] The temperature difference input sensing module, based on the grease pretreatment process, uses the timestamp sequence to perform hourly difference calculations on the heat exchange fluid inlet temperature difference and the external temperature difference. It determines whether there is a positive trend relationship between the unit heat load change and the temperature difference, and generates a temperature difference coupling association status identifier.

[0008] The energy consumption fluctuation extraction module constructs an energy consumption difference sequence based on the temperature difference coupling association state identifier according to the equipment category, calculates the difference between the maximum and minimum values ​​in the sequence, identifies the equipment section with excessive limits, and generates periodic energy consumption abnormal section information;

[0009] The abnormality attribution identification module calculates the ratio of the operating duration to the number of starts and stops based on the information of the abnormal energy consumption section of the cycle, and multiplies it by the energy consumption per unit time. If the ratio is lower than the preset standard and the product value is higher than the standard rated load of similar equipment, it is determined to be an unstable induced abnormality and a section-level energy consumption attribution structure map is generated;

[0010] The single-point mutation screening module calculates the unit material power consumption deviation rate for each device based on the section-level energy consumption attribution structure map. If the deviation rate is greater than the abnormal power consumption deviation threshold, it is marked as a jump node and the abnormal device type is attached to generate a device-level single-point energy consumption jump identification set.

[0011] The present invention has been improved in that the temperature difference coupling association state identifier includes the temperature difference value interval characteristics, the unit heat load coupling coefficient and the heat exchange equipment response category, the periodic energy consumption abnormal section information includes the abnormal equipment number, the energy consumption fluctuation amplitude index and the section time identifier, the section-level energy consumption attribution structure map specifically includes the equipment energy consumption inducement type, the section to which the attribution equipment belongs and the attribution judgment strength level, and the equipment-level single-point energy consumption jump identification set specifically refers to the jump node number, the jump offset degree classification and the jump corresponding processing process stage.

[0012] The present invention is improved in that the temperature difference input sensing module includes:

[0013] The thermal environment temperature difference analysis submodule extracts the inlet temperature, outlet temperature, heating medium temperature, and ambient air temperature of the heat exchange device based on the grease pretreatment process. It collects real-time temperature data at the inlet and outlet of the heat exchange fluid of the heat exchange device and records the corresponding timestamp sequence. It then calls the ambient air temperature value at the corresponding moment and performs hourly difference calculation with the heat exchange fluid inlet temperature to generate a thermal environment temperature difference sequence.

[0014] The temperature difference matching trend submodule calls the thermal environment temperature difference value sequence, synchronously calls the heat exchange value of the heat exchange device and the equipment heat load record value per unit time, performs a one-to-one mapping between the heat load value and the temperature difference difference sequence under the corresponding time series, calculates the matching coefficient, establishes a dual-sequence response trend between the heat exchange driving parameters and the unit heat load change, and obtains the heat load temperature difference matching trend sequence;

[0015] The temperature difference coupling correlation analysis submodule is based on the heat load temperature difference matching trend sequence. It determines the sign consistency between the change direction of the temperature difference value and the change direction of the unit heat load in each time period. If the two directions are consistent in consecutive cycles and the matching trend slope is greater than the trend reference slope value, it is marked as a coupling section and the equipment number and coupling section time period information are collected to generate a temperature difference coupling correlation status identifier.

[0016] The present invention is improved in that the energy consumption fluctuation extraction module includes:

[0017] The data extraction submodule extracts the unit output steam consumption, unit output electricity consumption, and unit output water consumption data of the pretreatment tank, heat exchange device, and separation centrifuge in the oil refining section based on the upper temperature difference coupling association state identifier. The unit energy consumption values ​​collected for each device in three consecutive process cycles are aggregated using a time series structure. The three types of energy consumption sequences for steam, electricity, and water are generated according to the type of equipment, and the unit energy consumption sequence set of the equipment category is obtained.

[0018] The difference calculation submodule calls the unit energy consumption sequence set of the equipment category, identifies the maximum and minimum values ​​in each category of sequence, performs difference calculation on the fluctuation amplitude of the unit energy consumption sequence of each equipment, and extracts the section with the largest fluctuation in consecutive cycles in the sequence. The total difference is constructed by equipment, and an energy consumption fluctuation intensity array is constructed based on the equipment category archive. The energy consumption fluctuation index value of the equipment is obtained by calculation.

[0019] The over-limit identification submodule reads the allowable fluctuation limit of the heat load corresponding to the equipment type according to the energy consumption fluctuation index value of the equipment, compares whether the fluctuation index value exceeds the corresponding limit, marks the exceeding equipment, and associates the corresponding time period, the equipment number and the process section information, establishes an abnormal section index list, and generates periodic energy consumption abnormal section information.

[0020] The present invention is improved in that the abnormal attribution identification module includes:

[0021] The equipment data aggregation submodule obtains the working duration, start-stop times, and total energy consumption per unit time of the heat exchange device and the material conveying device based on the abnormal energy consumption section information of the cycle, and classifies them according to the equipment number. The classified data structure is segmented and integrated on the time axis. Each parameter value of the equipment in the same cycle is combined into an equipment time period data group to generate an equipment cycle operation data group set;

[0022] The load structure calculation submodule calculates the ratio of the operation duration to the number of starts and stops for each set of data based on the equipment periodic operation data set, multiplies the ratio by the total energy consumption per unit time, constructs a sequence of equipment operation stability load values, performs a difference judgment on the result with the standard rated load of similar equipment, filters out data segments that do not meet the operation stability standard and have high load values, and obtains a sequence of unstable load characteristic values;

[0023] The section abnormality classification submodule obtains the corresponding grease processing section name based on the unstable load characteristic value sequence and the equipment number, binds the data segment marked as unstable load to the corresponding section identifier, maps the classification items according to the abnormality intensity level, establishes a section number and abnormality category comparison matrix, and generates a section-level energy consumption attribution structure map.

[0024] The present invention is improved in that the single point mutation screening module comprises:

[0025] Based on the section-level energy consumption attribution structure map, the operation data extraction submodule retrieves the operation cycle length, material handling volume, and unit cycle power consumption of the stirring device, degumming pump, and vacuum drying equipment in the refining section, and classifies them by equipment number. The classified data is horizontally integrated according to the time period structure to obtain the operation behavior data combination within the period corresponding to the equipment number, thereby obtaining the equipment cycle operation information set;

[0026] The power consumption offset calculation submodule calculates the unit power consumption ratio corresponding to the unit material processing volume for each device based on the device cycle operation information set, extracts the unit material power consumption value of the device in each cycle, calculates and obtains the unit material power consumption offset value of each device in each cycle, filters the device cycle data whose offset value is greater than the power consumption abnormal offset threshold, and generates a power consumption offset mutation structure array;

[0027] The abnormal node identification submodule calls the power consumption offset mutation structure array, screens the equipment number and the corresponding process segment identifier, and identifies the data segment in the offset value sequence where the consecutive offset nodes are greater than the threshold, adds the equipment type and the abnormality type, completes the binding of the three elements of equipment number, cycle position and abnormal label, and generates a device-level single-point energy consumption jump identifier set.

[0028] The present invention is improved in that the system further comprises:

[0029] The partition warning module calls the equipment-level single-point energy consumption jump identification set, calibrates the work section of the grease processing stage where the equipment is located, converts the ratio of the total number of equipment in each work section to the number of abnormal jump equipment, calculates the jump concentration, matches the concentration with the risk classification value, obtains the risk level corresponding to the work section, and generates the energy consumption monitoring partition warning level label;

[0030] The energy consumption monitoring zone warning level label includes the work section risk level number, equipment abnormality concentration ratio level and process link abnormality label.

[0031] The present invention is improved in that the partition warning module includes:

[0032] The partition calibration submodule calls the device-level single-point energy consumption jump identification set, identifies the process stage corresponding to each abnormal device, extracts the processing flow identification and section structure number, establishes the section partition identification index set based on the mapping result between the device number and the processing stage, and generates the equipment segmentation attribution structure information set;

[0033] The jump ratio calculation submodule counts the total number of devices and the number of abnormal devices in each section according to the equipment segmentation affiliation structure information set, calculates and obtains the corrected jump concentration value of each section, and generates a section abnormal concentration parameter set;

[0034] The risk level matching submodule calls the section abnormal concentration parameter set, compares the jump concentration value of each section with the energy consumption abnormal risk grading standard table, assigns the corresponding grading level identifier after comparison to the corresponding section number, generates a risk level label list, and obtains the energy consumption monitoring partition warning level label.

[0035] Compared with the prior art, the advantages and positive effects of the present invention are:

[0036] In the present invention, the accuracy and efficiency of energy utilization are improved through the fine monitoring and analysis of temperature differences, energy consumption fluctuations, abnormal states, single-point mutations and zoning warnings during the oil processing process. In the use of temperature difference data, by real-time monitoring of the difference between the temperature changes of the heat exchange fluid and the external temperature, the efficiency of thermal energy use can be more accurately analyzed, and an immediate response to the trend of thermal energy use can be achieved. Through in-depth analysis of energy consumption data in continuous time periods, specific sections of energy consumption anomalies can be identified, effectively preventing energy waste and the risk of equipment overload. By analyzing the relationship between the energy consumption of equipment operation and the working mode, data support is provided for equipment maintenance and energy saving. By recording and analyzing the specific data of the abnormal sections in detail, energy management personnel can quickly locate the cause of the problem, shorten the response time, and formulate more targeted improvement measures. The monitoring of single-point mutations refines the monitoring granularity, so that the energy consumption of each key equipment is strictly monitored, reducing the chain reaction risk brought about by energy consumption anomalies, ensuring the stability of the production process, and reducing energy costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 is a system flow chart of the present invention;

[0038] Figure 2 This is a flow chart of the temperature difference input sensing module of the present invention;

[0039] Figure 3 This is a flow chart of the energy consumption fluctuation extraction module of the present invention;

[0040] Figure 4 This is a flow chart of the abnormal attribution identification module of the present invention;

[0041] Figure 5 This is a flow chart of the single point mutation screening module of the present invention;

[0042] Figure 6 This is a flow chart of the partition warning module of the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0044] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings and are only for the convenience of describing the present invention and simplifying the description. They do not indicate or imply that the devices or elements referred to must have a specific direction, be constructed and operate in a specific direction, and therefore should not be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0045] See also Figure 1 The present invention provides a technical solution: an oil processing energy consumption monitoring system, the system comprising:

[0046] The temperature difference input sensing module extracts the inlet temperature, outlet temperature, heating medium temperature, and ambient air temperature of the heat exchange device based on the grease pretreatment process. It then uses a timestamp sequence to perform hourly difference calculations on the inlet temperature difference of the heat exchange fluid and the external temperature difference. It then performs a double-sequence matching analysis on the difference and the unit heat load of the corresponding time period to determine whether there is a positive trend relationship between the unit heat load change and the temperature difference, and generates a temperature difference coupling association status identifier.

[0047] The energy consumption fluctuation extraction module, based on the temperature difference coupling association state identification, extracts the unit output steam consumption, unit output electricity consumption, and unit output water consumption of the oil refining process equipment for three consecutive time periods. It then constructs an energy consumption difference sequence by equipment type and calculates the difference between the maximum and minimum fluctuation amplitudes in each sequence. The difference result is compared with the set allowable fluctuation limit of the heat load to identify equipment sections with excessive fluctuations and generate information on periodic energy consumption abnormalities.

[0048] Based on the information of the abnormal periodic energy consumption section, the abnormal attribution identification module obtains the operating duration, start-stop times, and total energy consumption per unit time of the heat exchange device and material conveying device in the abnormal section. Each data item is aggregated by equipment number, and the ratio of operating duration to start-stop times is calculated. The product is then multiplied by the energy consumption per unit time. If the ratio is lower than the preset standard and the product value is higher than the standard rated load of similar equipment, it is determined to be an unstable induced abnormality and classified into the corresponding oil and fat processing section according to the equipment number, and a section-level energy consumption attribution structure map is generated. The ratio of operating duration to start-stop times is an important factor in measuring the stability of equipment operation. The product and energy consumption are used to determine load consistency. The reference value limit can be set based on the upper and lower limits of the rated load specified in the equipment manual.

[0049] The single-point mutation screening module uses the section-level energy consumption attribution structure map to retrieve the operating cycle length, material handling volume, and unit power consumption of the stirring device, degumming pump, and vacuum drying equipment in the refining section. It calculates the unit material power consumption deviation rate for each device and performs difference normalization with the average value of similar devices. If the deviation rate exceeds the abnormal power consumption deviation threshold, it is marked as a jump node and the abnormal device type is attached, generating a device-level single-point energy consumption jump identification set.

[0050] The unit material power consumption deviation rate refers to the difference between the actual unit material processing power consumption and the historical equipment average;

[0051] The partition warning module calls the equipment-level single-point energy consumption jump identification set to calibrate the work section of the grease treatment stage where the equipment is located. It calculates the jump concentration based on the ratio of the total number of equipment in each work section to the number of equipment with abnormal jumps, matches the concentration with the risk classification value, obtains the risk level corresponding to the work section, and generates the energy consumption monitoring partition warning level label;

[0052] The temperature difference coupling association status identification includes the temperature difference range characteristics, unit heat load coupling coefficient and heat exchange equipment response category. The periodic energy consumption abnormal section information includes the abnormal equipment number, energy consumption fluctuation amplitude index and section time identification. The section-level energy consumption attribution structure map specifically includes the equipment energy consumption inducement type, the section to which the attribution equipment belongs and the attribution judgment strength level. The equipment-level single-point energy consumption jump identification set specifically refers to the jump node number, jump offset degree classification and jump corresponding processing stage. The energy consumption monitoring partition warning level label includes the section risk level number, equipment abnormality concentration ratio level and process link abnormality label.

[0053] See also Figure 2 , the temperature difference input sensing module includes:

[0054] The thermal environment temperature difference analysis submodule extracts the inlet temperature, outlet temperature, heating medium temperature, and ambient air temperature of the heat exchange device based on the grease pretreatment process. It collects real-time temperature data at the inlet and outlet of the heat exchange fluid of the heat exchange device and records the corresponding timestamp sequence. It then calls the ambient air temperature value at the corresponding moment and performs hourly difference calculation with the heat exchange fluid inlet temperature to generate a thermal environment temperature difference sequence.

[0055] Obtain the inlet temperature, outlet temperature, heating medium temperature and ambient air temperature of the heat exchange device in the oil pretreatment section. In the process of collecting the real-time temperature data of the inlet and outlet of the heat exchange fluid, a high-precision thermocouple temperature sensor can be installed on each heat exchange device, and a data collector with a time stamp recording function can be configured to record the temperature data once per minute. The ambient air temperature at the corresponding moment is measured by the ambient temperature sensor in the adjacent area. For example, in a typical sample, if the inlet temperature of the heat exchange fluid is 92.6℃, the outlet temperature is 68.1℃, the ambient temperature is 36.4℃, and the heating medium temperature is 115.2℃ at a certain moment, then the moment is recorded as "2025-04-25 At 09:15:00, the difference between the inlet temperature of 92.6°C and the ambient temperature of 36.4°C is calculated to be 56.2°C. This difference is bound to the timestamp and arranged in chronological order to form a thermal environment temperature difference sequence. For the "collection" and "calculation" operations in the above process, the execution logic is as follows: the collected temperature value is converted into an analog voltage signal and converted into a numerical record in real time by the A / D conversion unit. For the "difference calculation" process, the inlet temperature and the ambient temperature are subtracted point by point to form a difference sequence. The data sequence is set as follows: , , then the calculated difference sequence is The timestamps “09:15”, “09:16” and “09:17” are used to mark the thermal environment temperature difference value sequence, which constitutes the basic input for subsequent temperature difference trend judgment and generates the thermal environment temperature difference sequence.

[0056] The temperature difference matching trend submodule calls the thermal environment temperature difference sequence, synchronously retrieves the heat exchange value of the heat exchange device and the equipment heat load record value per unit time, performs a one-to-one mapping between the heat load value and the temperature difference sequence under the corresponding time series, calculates the matching coefficient, establishes a dual-sequence response trend between the heat exchange drive parameters and the unit heat load change, and obtains the heat load temperature difference matching trend sequence;

[0057] Call the thermal environment temperature difference sequence, and synchronously call the heat exchanger heat exchange unit and equipment heat load record value per unit time. It is necessary to ensure that the above three types of data have time stamp consistency. The heat exchange per unit time can be obtained by measuring the inlet and outlet temperature difference of the heat medium and combining it with the flow data. The heat load value is provided by the heat exchanger control system. The sampling time interval is set to 1 minute. The calculation formula for that minute is:

[0058] ;

[0059] in is the specific heat capacity, taking water as 4.18 kJ / kg·K, Take 1 kg / L as the density, The flow rate is set to 3.2 L / min. ,but , the equipment heat load at this moment is recorded as 340.5kJ. According to the temperature difference at this time point of 56.2℃, a one-to-one mapping matching is performed, and the three values ​​are connected in parallel as data nodes. At the same time, the next cycle node is recorded, and all nodes are arranged in chronological order. The matching coefficient between the heat load and the temperature difference sequence is calculated through the linear correlation function, using the Pearson correlation coefficient formula:

[0060] ;

[0061] in is the temperature difference series, For the heat load sequence, enter the values ​​at three time points: , , calculated , indicating that the two are strongly linearly correlated. Then, according to the time sequence, the temperature difference value at each time point and the corresponding unit heat load value are established into data pairs according to the corresponding relationship to generate a heat load temperature difference matching trend sequence.

[0062] The temperature difference coupling correlation analysis submodule is based on the heat load temperature difference matching trend sequence. It determines the sign consistency between the change direction of the temperature difference value and the change direction of the unit heat load in each period. If the two directions are consistent in consecutive cycles and the matching trend slope is greater than the trend reference slope value, it is marked as a coupling section and the equipment number and coupling section time period information are collected to generate the temperature difference coupling correlation status identification.

[0063] Based on the heat load temperature difference matching trend sequence, the sign consistency of the change direction of the temperature difference value in each period and the change direction of the unit heat load is judged. The sign of the difference change needs to be judged according to the data of two adjacent moments in the sequence. For example, the temperature difference sequence is , the heat load is , the change directions are negative, negative, and negative respectively, so the two have consistent directions. At the same time, it is necessary to judge whether the trend slope exceeds the trend reference slope value. The trend slope is linearly fitted. The base slope value is calculated It is a preset judgment standard for the system, used to identify the difference in trend intensity. The setting basis is the average value of the minimum linear fitting slope value between the normal working condition heat load and temperature difference in multiple batches of debugging results. The actual setting is In actual operation, 10 sets of working condition data were collected and linear regression was performed to extract the slope. The slope range was 1.65 to 2.03, so the median was trimmed down to 1.85 as the stability judgment benchmark of the trend response. If the fitting trend is , put the above three points into calculation, and the fitting slope is , then the trend of this section meets the coupling standard, and these three time periods are marked as coupling sections. The corresponding device number and the corresponding timestamp interval "2025-04-25 09:15:00" to "2025-04-25 09:17:00" are called, the coupling label is recorded, and a dual matching dictionary structure of device number and coupling time period is constructed to finally generate the temperature difference coupling association state identifier.

[0064] See also Figure 3 , the energy consumption fluctuation extraction module includes:

[0065] The data extraction submodule extracts the unit output steam consumption, unit output electricity consumption, and unit output water consumption data of the pretreatment tank, heat exchange device, and separation centrifuge in the oil refining section based on the upper temperature difference coupling association state identifier. The unit energy consumption values ​​collected for each device in three consecutive process cycles are aggregated using a time series structure. The three types of energy consumption sequences for steam, electricity, and water are generated according to the type of equipment, and the unit energy consumption sequence set of the equipment category is obtained.

[0066] Based on the temperature difference coupling association state identification, the unit output steam consumption, unit output electricity consumption and unit output water consumption data of the pretreatment tank, heat exchange device and separation centrifuge in the oil refining section are extracted. First, the output benchmark of each type of equipment in three consecutive production cycles must be determined and normalized in units of "tons of oil". When obtaining the real-time energy consumption records of the equipment, the steam consumption can be directly recorded by the flow meter at the steam flow rate and the cumulative value can be obtained by time integration. The electricity consumption is determined by the difference in the meter readings, and the water consumption is provided by the cumulative value of the water flow meter. The normalization method is to divide the energy consumption per cycle by the output tons of the corresponding cycle to form a unit energy consumption value. In this example, it is assumed that the heat exchange device consumes 210 kg of steam in a certain cycle and produces 2 tons of oil. The unit steam consumption is 105 kg / t, the electricity consumption is set to 132 kWh, the unit electricity consumption is 66 kWh / t, the water consumption is 920L, and the unit water consumption is 460 L / t, extract the above indicators in three consecutive cycles, combine the unit energy consumption values ​​of the three types of equipment in the three cycles to form the original energy consumption array, and divide it into three sequence structures of steam, electricity, and water according to the equipment category. After alignment based on the timestamp sequence, the structure is as follows:

[0067] Table 1 Sample table of unit energy consumption time series

[0068]

[0069] As shown in Table 1, the unit energy consumption data are organized into a time series, which is then packaged into a structured array according to the equipment category to generate a unit energy consumption series set by equipment category.

[0070] The difference calculation submodule calls the unit energy consumption sequence set of equipment categories, identifies the maximum and minimum values ​​in each sequence, calculates the difference of the fluctuation amplitude of each equipment unit energy consumption sequence, and extracts the maximum fluctuation segment of the continuous cycle in the sequence. The total difference is constructed by equipment, and the energy consumption fluctuation intensity array is constructed according to the equipment category archive. The formula is:

[0071] ;

[0072] Calculate and obtain the energy consumption fluctuation index value of the equipment;

[0073] in, Indicates the Energy consumption fluctuation index value of this type of equipment, 、 Respectively represent The maximum and minimum normalized values ​​of the unit output energy consumption of this type of equipment, Indicates the Class 1 equipment The normalized value of energy consumption per unit output in a cycle, Indicates the The normalized mean value of the unit output energy consumption of this type of equipment in all cycles, Indicates the number of time periods being monitored;

[0074] Call the equipment category unit energy consumption sequence set, first extract the maximum and minimum values ​​in each sequence, for example, the unit steam consumption sequence of the heat exchange device is [105, 113, 97] kg / t, where the maximum value is , minimum , the fluctuation range is the absolute difference , then perform standard deviation calculation on the unit energy consumption series of each equipment to determine the severity of fluctuations, and introduce the periodic value of normalized unit output energy consumption into the standard deviation part for processing. The formula is as follows: ;

[0075] The formula parameters are described as follows:

[0076] : No. The energy consumption fluctuation index value of this type of equipment indicates the degree of fluctuation of unit output energy consumption over multiple cycles;

[0077] 、 :Respectively The maximum and minimum values ​​of unit output energy consumption of the equipment in the monitoring period, all of which are normalized unit values;

[0078] : No. Class 1 equipment Normalized value of energy consumption per unit output per cycle, : No. The average normalized value of unit energy consumption of this type of equipment;

[0079] : The number of statistical time periods is set to 3.

[0080] Take the steam sequence as an example: [105, 113, 97], its average value is , the sum of squared differences is , the standard deviation is , the final volatility index is: , indicating that the energy consumption fluctuation level of this equipment type in this sequence is 104.48.

[0081] The over-limit identification submodule reads the allowable fluctuation limit of the heat load corresponding to the equipment type based on the energy consumption fluctuation index value of the equipment, compares the fluctuation index value to see if it exceeds the corresponding limit, marks the equipment that exceeds the limit, and associates the corresponding time period, equipment number and process section information to establish an abnormal section index list and generate period energy consumption abnormal section information;

[0082] Based on the energy consumption fluctuation index value of the equipment, the preset allowable fluctuation limit of the heat load for each type of equipment is read. The limit is set based on the allowable range of load fluctuations for typical industrial equipment in the national standard GB / T 2589-2020. The limit is set in combination with the fluctuation tolerance design of the equipment category. For example, for the heat exchange device, the allowable limit of heat load fluctuation is set to 92.00 kJ / t. If the calculated fluctuation index value is 104.48, it exceeds the limit and is identified as an abnormality. The equipment number is set to HXZ-02, and the corresponding period is "09:30–10:00", which belongs to the deacidification section. Then, an identification key-value pair is generated in the abnormal record structure, including the equipment number, time period, and the name of the section to which it belongs. Finally, an abnormal section index list is constructed to generate the periodic energy consumption abnormal section information.

[0083] The process of setting this value is as follows: Under preset conditions, multiple batches of operating samples of similar equipment are measured, and the thermal load fluctuation ranges under full load, half load and variable load operation are counted. The 10 groups of samples are obtained, namely: 88.6, 91.2, 92.3, 90.5, 93.4, 89.7, 91.6, 90.1, 92.8, and 89.9. The arithmetic mean is taken as the setting reference value, which is calculated as , combined with the equipment safety margin, the weighted adjustment is set to 92.00 kJ / t. This result shows that 104.48 > 92.00, indicating that the current heat exchanger energy consumption fluctuation range exceeds the stable operating range and requires the abnormal identification process.

[0084] See also Figure 4 , the abnormal attribution identification module includes:

[0085] The equipment data aggregation submodule obtains the operating duration, start-stop times, and total energy consumption per unit time of the heat exchanger and material conveying device based on the abnormal periodic energy consumption section information, and classifies them by device number. The classified data structure is then segmented and integrated along the time axis. Each parameter value of the equipment in the same period is grouped into an equipment time period data group to generate an equipment period operation data group set.

[0086] To obtain the operating duration, start-stop frequency, and total energy consumption per unit time of the heat exchanger and material conveying device in the abnormal periodic energy consumption segment information, the energy consumption monitoring records and operation logs of the corresponding devices must be retrieved respectively. The operating duration can be calculated by the difference in the time points of the equipment status change. For example, if a heat exchanger starts operating at 08:00 and stops at 08:45, the operating duration is 45 minutes. The start-stop frequency is obtained by counting the total number of start-up and shutdown signals within the period. The total energy consumption per unit time is obtained by integrating the energy consumption accumulation curve within the period. In the system, all records are classified by equipment number. For example, equipment HXZ-01 corresponds to three groups of data: [45 min, 3 times, 12.8 kWh], [43 min, 2 times, 11.6 kWh], and [46 min, 3 times, 13.2 kWh]. The classified data is segmented and integrated on the time axis. The operating duration, start-stop frequency, and total energy consumption in each time period are aggregated into a single record. These records are then sequentially organized into an equipment operating status record matrix, which in turn forms the equipment periodic operation data set.

[0087] The load structure calculation submodule calculates the ratio of the equipment's periodic operation data set to the number of starts and stops for each set of data. This ratio is multiplied by the total energy consumption per unit time to construct a sequence of equipment operation stability load values. The result is then compared with the standard rated load of similar equipment to determine the difference. Data segments that do not meet the operation stability standards and have high load values ​​are screened out to obtain a sequence of unstable load characteristic values.

[0088] Based on the equipment cycle operation data set, the ratio of operation duration to start and stop times is calculated for each set of data. If a device runs for 45 minutes in a cycle and starts and stops 3 times, the ratio is 15. If the total energy consumption per unit time of the cycle is 12.8 kWh, the operation stability load value is , the process is performed once for each data segment, and the sequence is obtained , construct the load value sequence of the equipment operation stability, and make a difference judgment between the sequence and the standard rated load of similar equipment. The standard rated load value is set to 200, which is obtained by normalizing the rated operating energy consumption of the equipment manual. The rated continuous operation power consumption is 13.3 kWh, the average start and stop times are 2.5 times, and the standard time ratio is 14 minutes. The set load is , adjust the normalization coefficient to 2.68 times and round it to 200. If the load value is greater than this value and the ratio is less than 12, it is determined to be an unstable operation segment. On this basis, the segment data markers with a value greater than 200 and a corresponding ratio less than 12 in the sequence are screened out, and their time position and equipment number are extracted to obtain the unstable load characteristic value sequence.

[0089] The section abnormality classification submodule obtains the corresponding grease treatment section name based on the unstable load characteristic value sequence and the equipment number, binds the data segment marked as unstable load to the corresponding section identifier, and maps the classification items according to the abnormal intensity level. It establishes a comparison matrix between the section number and the abnormality category, and generates a section-level energy consumption attribution structure map;

[0090] According to the sequence of unstable load characteristic values, the source is traced upward by the equipment number, and the section identifier registered in the system deployment table for the equipment number is retrieved. For example, if the number HXZ-01 corresponds to the pretreatment section in the table, the mark of the equipment in the abnormal time period "09:00-09:30" is bound to the "pretreatment section". At the same time, all abnormal mark entries are classified and sorted by section, and abnormal intensity level mapping is performed. The level classification rule is: if the proportion of abnormal number to the total number of equipment in the section is within 0-20%, it is marked as L1; within 21%-50% is L2; ​​and more than 50% is L3. Suppose there are 6 equipment in a section, 3 of which are marked as abnormal, then the level is L2. Create a mapping item {"pretreatment section": L2}. In this way, all abnormal sections are uniformly generated to generate a double-field structure map record entry "section number + abnormal level". The results are aggregated to obtain the section-level energy consumption attribution structure map.

[0091] See also Figure 5 , the single point mutation screening module includes:

[0092] The operation data extraction submodule retrieves the operation cycle length, material handling volume, and unit cycle power consumption of the agitator, degumming pump, and vacuum drying equipment in the refining section based on the section-level energy consumption attribution structure map. It then categorizes the equipment by equipment number and integrates the categorized data horizontally according to the time period structure to obtain the combination of operation behavior data within the period corresponding to the equipment number, thereby obtaining the equipment cycle operation information set.

[0093] Based on the section-level energy consumption attribution structure, the system retrieves the operating cycle length, material handling volume, and power consumption per cycle for the agitator, degumming pump, and vacuum drying equipment in the refining section. The system searches the equipment operation logs and material transfer records based on the equipment number, extracting data within the corresponding time window using the cycle length as the query criterion. The equipment operating cycle length is determined by the time difference between the equipment's start and stop signals. For example, if a agitator starts at 09:00 and stops at 09:45, the cycle length is 45 minutes. The material handling volume is collected by an online weighing system or flow meter. The power consumption per cycle is normalized by the energy consumption monitoring module by summarizing the energy consumption readings within the cycle and calculating the value based on the output. For example, agitator SP-01 processed 3.6 tons of material during its operating cycle from 09:00 to 09:45, with a total energy consumption of 16.2 kWh. This translates to a power consumption per cycle of 4.5 kWh / t. The above data is classified by equipment number in the system and organized into a structured data group in a time series format. During the horizontal integration process, the operating cycle length, material handling volume, and unit energy consumption value of each equipment in each cycle are combined into a set of three-dimensional data structures. After organization, the equipment cycle operation information set is formed, for example, as shown below:

[0094] Table 2 Sample table of equipment operation information data

[0095]

[0096] As shown in Table 2, each set of data is organized by time period, and the three core indicators are updated in real time through the data collection system.

[0097] The power consumption offset calculation submodule calculates the unit power consumption ratio corresponding to the unit material processing volume of each device based on the equipment cycle operation information set, and extracts the unit material power consumption value of the device in each cycle using the formula:

[0098] ;

[0099] Obtain the unit material power consumption offset value of each device in each cycle through calculation, filter the device cycle data whose offset value is greater than the abnormal power consumption offset threshold, and generate a power consumption offset mutation structure array;

[0100] in, Indicates the The device in The unit material power consumption offset value of each cycle, Indicates the Device No. The normalized value of energy consumption per cycle, Indicates the Device No. Normalized value of material handling volume per cycle, is the number of time cycles for similar devices, is the device type adjustment factor, For the Device No. Cycle running time, For the The total number of starts and stops of each device in all cycles;

[0101] Based on the equipment cycle operation information set, the unit power consumption ratio corresponding to the unit material processing volume is calculated for each device. After extracting the unit material power consumption of each cycle, the difference ratio relative to the average unit power consumption of the equipment category is calculated. The power consumption offset value is then calculated by combining the amplitude modulation factor. Taking the equipment SP-01 as an example, its unit power consumption values ​​for the three cycles are 4.5, 4.82, and 4.35 kWh / t respectively, with an average value of kWh / t, for the second cycle, the unit material power consumption offset is , set the device type adjustment factor , running time , the total number of starts and stops , then use the formula:

[0102] ;

[0103] Enter the value: If the system sets the abnormal power consumption offset threshold to 0.28 (this value is set by the boundary value of the normal offset range of more than 90% of similar devices), then this cycle offset value is greater than the threshold and is determined to be an abnormal offset segment. Finally, the cycle data with offset values ​​greater than the threshold is screened and collected, and summarized into a power consumption offset mutation structure array.

[0104] The abnormal node identification submodule calls the power consumption offset mutation structure array, filters the device number and the corresponding process segment identifier, and identifies the data segments in the offset value sequence where the consecutive offset nodes are greater than the threshold. It then adds the device type and abnormality type, completes the binding of the three elements of device number, cycle position, and abnormality label, and generates a device-level single-point energy consumption jump identifier set.

[0105] The power consumption offset mutation structure array is called. First, based on the device number bound to each offset record, the corresponding process segment information in the device configuration list is retrieved. For example, number SP-01 corresponds to the debonding segment. The device cycle segment identified as abnormal is sequentially extracted with the number, cycle time, offset value, and device type tag. The offset value sequence is then identified to determine whether there are two or more consecutive abnormal cycles. If so, the offset value is marked as "continuous offset"; otherwise, it is marked as "isolated offset." For example, if a device has two consecutive cycle offset values ​​of 0.35 and 0.31, both exceeding the threshold of 0.28, it is considered a continuous offset segment. These results are then bound to the device number, cycle position, and anomaly tag, represented as a triplet (SP-01, 10:00–11:30, continuous offset). Finally, the data is output as a list, generating a device-level single-point energy consumption jump identifier set.

[0106] See also Figure 6 , the partition warning module includes:

[0107] The partition calibration submodule calls the equipment-level single-point energy consumption jump identification set to identify the process stage corresponding to each abnormal equipment, extracts the processing flow identification and section structure number, establishes the section partition identification index set based on the mapping result between the equipment number and the processing stage, and generates the equipment segmentation affiliation structure information set;

[0108] When invoking the device-level single-point energy consumption transition identification set, the abnormal device number, bound cycle, and device type tag are extracted from the record. The device number is then matched against the device information database to obtain the corresponding process identifier (e.g., "degumming," "drying") and process section structure number (e.g., "DGS-03"). The process stage and structure number are then bound together as a mapping pair, using the device number as the primary key. For example, for device SP-02, its bound tag is "vacuum drying equipment." The system queries the equipment list and finds its associated process flow as "drying section" and structure number as "DRY-02." The device record is structured as (SP-02, drying section, DRY-02). This process identifier is then categorized for all abnormal devices. Finally, a structure item set is generated, which is used for subsequent attribution partitioning and transition intensity calculations, forming the device segment attribution structure information set.

[0109] The jump ratio calculation submodule counts the total number of devices and the number of abnormal devices in each section according to the equipment segmentation structure information set, using the formula:

[0110] ;

[0111] Calculate and obtain the corrected jump concentration value of each section, and generate the section abnormal concentration parameter set;

[0112] in, Indicates the Corrected jump concentration value of the section, For the The number of equipment marked as abnormal in the section, For the Total number of equipment in the section, For the The number of corresponding processing cycles in the section, For the Equipment layout concentration coefficient in the work section;

[0113] According to the equipment segmentation structure information set, count the total number of equipment in each section Number of devices with abnormal jumps , and calculate its basic jump ratio , introducing the processing cycle number factor and equipment concentration factor Dynamic corrections are made. The equipment concentration factor is reflected by the number of devices per unit area in the work section layout, and the number of processing cycles is measured as the sum of the number of cycles. In this example, there are 10 devices in the debonding section (number DG-01), 3 of which are marked as jump anomalies. The number of processing cycles is 24, and the equipment density concentration factor is 3.2 (i.e., 10 devices occupy 3.1 square meters of work area). The calculation process is as follows: ;

[0114] The corrected concentration of the degumming section jump is 3.828. If it is required to be rounded to two decimal places according to the rules, the value is 3.83. The result is written into the section abnormality record structure to form the section abnormality concentration parameter set.

[0115] The risk level matching submodule calls the section abnormal concentration parameter set, compares the jump concentration value of each section with the energy consumption abnormal risk grading standard table, assigns the corresponding grade identification after comparison to the corresponding section number, generates a risk level label list, and obtains the energy consumption monitoring partition warning level label;

[0116] Call the section abnormal concentration parameter set and map the corrected jump concentration value of each section to the preset energy consumption abnormal risk grading standard table. Assume that the risk standard is: For level L1, is L2, For example, the aforementioned debonding section has a jump concentration value of 3.83, corresponding to a risk level of L3. The section number DG-01 is associated with its level L3, creating a structure record (DG-01, L3). All matching results are organized into a list, forming energy consumption monitoring zone warning level labels. Each label record indicates the energy consumption status of the zone and supports subsequent configuration and analysis of the management system.

[0117] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. Oil processing energy consumption monitoring system, characterized by: The system comprises: The temperature difference input sensing module, based on the grease pretreatment process, uses the timestamp sequence to perform hourly difference calculations on the heat exchange fluid inlet temperature difference and the external temperature difference. It determines whether there is a positive trend relationship between the unit heat load change and the temperature difference, and generates a temperature difference coupling association status identifier. The temperature difference input sensing module includes: The thermal environment temperature difference analysis submodule extracts the inlet temperature, outlet temperature, heating medium temperature, and ambient air temperature of the heat exchange device based on the grease pretreatment process. It collects real-time temperature data at the inlet and outlet of the heat exchange fluid of the heat exchange device and records the corresponding timestamp sequence. It then calls the ambient air temperature value at the corresponding moment and performs hourly difference calculation with the heat exchange fluid inlet temperature to generate a thermal environment temperature difference sequence. The temperature difference matching trend submodule calls the thermal environment temperature difference value sequence, synchronously calls the heat exchange value of the heat exchange device and the equipment heat load record value per unit time, performs a one-to-one mapping between the heat load value and the temperature difference difference sequence under the corresponding time series, calculates the matching coefficient, establishes a dual-sequence response trend between the heat exchange driving parameters and the unit heat load change, and obtains the heat load temperature difference matching trend sequence; The temperature difference coupling correlation analysis submodule is based on the heat load temperature difference matching trend sequence. It judges the symbol consistency of the change direction of the temperature difference value and the change direction of the unit heat load in each period. If the two directions are consistent in consecutive cycles and the matching trend slope is greater than the trend reference slope value, it is marked as a coupling section and the equipment number and coupling section time period information are collected to generate a temperature difference coupling correlation status identifier. The energy consumption fluctuation extraction module constructs an energy consumption difference sequence based on the temperature difference coupling association state identifier according to the equipment category, calculates the difference between the maximum and minimum values ​​in the sequence, identifies the equipment section with excessive limits, and generates periodic energy consumption abnormal section information; The energy consumption fluctuation extraction module includes: Based on the temperature difference coupling association state identifier, the data extraction submodule extracts the unit output steam consumption, unit output electricity consumption, and unit output water consumption data of the pretreatment tank, heat exchange device, and separation centrifuge in the oil refining section. The unit energy consumption values ​​collected for each device in three consecutive process cycles are aggregated using a time series structure. The three types of energy consumption sequences for steam, electricity, and water are generated by equipment type, resulting in a set of unit energy consumption sequences for each equipment type. The difference calculation submodule calls the unit energy consumption sequence set of the equipment category, identifies the maximum and minimum values ​​in each category of sequence, performs difference calculation on the fluctuation amplitude of the unit energy consumption sequence of each equipment, and extracts the section with the largest fluctuation in consecutive cycles in the sequence. The total difference is constructed by equipment, and an energy consumption fluctuation intensity array is constructed based on the equipment category archive. The energy consumption fluctuation index value of the equipment is obtained by calculation. The over-limit identification submodule reads the allowable fluctuation limit of the heat load corresponding to the equipment type based on the energy consumption fluctuation index value of the equipment, compares the fluctuation index value to see if it exceeds the corresponding limit, marks the equipment that exceeds the limit, and associates the corresponding time period, the equipment number and the process section information to establish an abnormal section index list and generate period energy consumption abnormal section information; The abnormality attribution identification module calculates the ratio of the operating duration to the number of starts and stops based on the information of the abnormal energy consumption section of the cycle, and multiplies it by the energy consumption per unit time. If the ratio is lower than the preset standard and the product value is higher than the standard rated load of similar equipment, it is determined to be an unstable induced abnormality and a section-level energy consumption attribution structure map is generated; The abnormal attribution identification module includes: The equipment data aggregation submodule obtains the working duration, start-stop times, and total energy consumption per unit time of the heat exchange device and the material conveying device based on the abnormal energy consumption section information of the cycle, and classifies them according to the equipment number. The classified data structure is segmented and integrated on the time axis. Each parameter value of the equipment in the same cycle is combined into an equipment time period data group to generate an equipment cycle operation data group set; The load structure calculation submodule calculates the ratio of the operation duration to the number of starts and stops for each set of data based on the equipment periodic operation data set, multiplies the ratio by the total energy consumption per unit time, constructs a sequence of equipment operation stability load values, performs a difference judgment on the result with the standard rated load of similar equipment, filters out data segments that do not meet the operation stability standard and have high load values, and obtains a sequence of unstable load characteristic values; The section abnormality classification submodule obtains the corresponding grease processing section name based on the unstable load characteristic value sequence and the equipment number, binds the data segment marked as unstable load to the corresponding section identifier, and maps the classification items according to the abnormal intensity level. It establishes a comparison matrix between the section number and the abnormality category, and generates a section-level energy consumption attribution structure map; The single-point mutation screening module calculates the unit material power consumption deviation rate for each device based on the section-level energy consumption attribution structure map. If the deviation rate is greater than the abnormal power consumption deviation threshold, it is marked as a jump node and the abnormal device type is attached to generate a device-level single-point energy consumption jump identification set.

2. The oil processing energy consumption monitoring system according to claim 1, characterized in that: The temperature difference coupling association state identifier includes the temperature difference interval characteristics, the unit heat load coupling coefficient and the heat exchange equipment response category. The periodic energy consumption abnormal section information includes the abnormal equipment number, the energy consumption fluctuation amplitude index and the section time identifier. The section-level energy consumption attribution structure map specifically includes the equipment energy consumption inducement type, the section to which the attribution equipment belongs and the attribution judgment strength level. The equipment-level single-point energy consumption jump identifier set specifically refers to the jump node number, the jump offset degree classification and the jump corresponding processing process stage.

3. The oil processing energy consumption monitoring system according to claim 2, characterized in that: The single point mutation screening module includes: Based on the section-level energy consumption attribution structure map, the operation data extraction submodule retrieves the operation cycle length, material handling volume, and unit cycle power consumption of the stirring device, degumming pump, and vacuum drying equipment in the refining section, and classifies them by equipment number. The classified data is horizontally integrated according to the time period structure to obtain the operation behavior data combination within the period corresponding to the equipment number, thereby obtaining the equipment cycle operation information set; The power consumption offset calculation submodule calculates the unit power consumption ratio corresponding to the unit material processing volume for each device based on the device cycle operation information set, extracts the unit material power consumption value of the device in each cycle, calculates and obtains the unit material power consumption offset value of each device in each cycle, filters the device cycle data whose offset value is greater than the power consumption abnormal offset threshold, and generates a power consumption offset mutation structure array; The abnormal node identification submodule calls the power consumption offset mutation structure array, screens the equipment number and the corresponding process segment identifier, and identifies the data segment in the offset value sequence where the consecutive offset nodes are greater than the threshold, adds the equipment type and the abnormality type, completes the binding of the three elements of equipment number, cycle position and abnormal label, and generates a device-level single-point energy consumption jump identifier set.

4. The oil processing energy consumption monitoring system according to claim 3, characterized in that: The system further comprises: The partition warning module calls the equipment-level single-point energy consumption jump identification set, calibrates the work section of the grease processing stage where the equipment is located, converts the ratio of the total number of equipment in each work section to the number of abnormal jump equipment, calculates the jump concentration, matches the concentration with the risk classification value, obtains the risk level corresponding to the work section, and generates the energy consumption monitoring partition warning level label; The energy consumption monitoring zone warning level label includes the work section risk level number, equipment abnormality concentration ratio level and process link abnormality label.

5. The oil processing energy consumption monitoring system according to claim 4, characterized in that: The partition warning module includes: The partition calibration submodule calls the device-level single-point energy consumption jump identification set, identifies the process stage corresponding to each abnormal device, extracts the processing flow identification and section structure number, establishes the section partition identification index set based on the mapping result between the device number and the processing stage, and generates the equipment segmentation attribution structure information set; The jump ratio calculation submodule counts the total number of devices and the number of abnormal devices in each section according to the equipment segmentation affiliation structure information set, calculates and obtains the corrected jump concentration value of each section, and generates a section abnormal concentration parameter set; The risk level matching submodule calls the section abnormal concentration parameter set, compares the jump concentration value of each section with the energy consumption abnormal risk grading standard table, assigns the corresponding grading level identifier after comparison to the corresponding section number, generates a risk level label list, and obtains the energy consumption monitoring partition warning level label.

Images