Monitoring and forecasting of equipment technical condition
Clover PMM collects data streams from telemetry sensors at oil and gas production, oil transportation and and oil refinery enterprises.
Clover PMM has mathematical models that implement engineering principles in analyzing the operation of equipment and external parameters.
Predicts failures and stops
The joint operation of math-models, machine learning and artificial intelligence allows predicting possible breakdowns and their causes.
Clover PMM Capabilities
Clover PMM analyzes large data streams. The system tracks anomalies in the operation of the equipment and predicts breakdowns with a planning horizon from 2 weeks
Breakdown prediction allows forestalling operations. Clover PMM customers in the oil and gas industry reduce their downtimes by an average of 35-45%
Effective predictions of failure or replacement reduce the cost of unplanned equipment repairs by 13-15%
Most businesses in the oil and gas industry rely on a predictive and preventive maintenance plan. To implement the predictive and preventive maintenance plan, the data is further analyzed by technologists.
But no person is able to analyze more than 7-10 parameters at a time. At the same time, just one compressor unit is able to give out 50 to 100 parameters of different distinctiveness.
effectively analyzes people
provides a standard compressor unit
3+ million parameters
can analyze Clover PMM
Clover PMM solution
Clover PMM incorporates two modules: search for anomalies and predictive analytics. The prediction is based on unique MX-models (Math eXperience). These are developed using physical and mathematical modeling, machine learning and expert-engineer rule packages.
Instead of a single unit, mathematical models describe a set of interrelated equipment. For example, instead of individual components, the MX model of the compressor unit from Clover describes the installation as a whole. Clover PMM models also take into account environmental parameters and properties of pumped environments.
Mathematical models are usually developed under "ideal" conditions. It is impossible to use them in controlling real production in the oil and gas industry; any changes in the parameters will require appropriate adjustments. And disparate unit models do not provide a uniform picture of the equipment complex. At the same time, production data is often "raw"– with errors in quality and quantity.
The Clover team came up with a solution to the math-model problem from two sides. The scientific part of the team studied the operation of the equipment of oil producing, processing and transporting enterprises in all modes, climatic zones, regulated and non-regulated situations. Meanwhile, IT professionals created a convenient multi-component platform on which the models work.
Mathematicians and engineers, production experts and programmers summarized the theory, general prediction methods, probabilistic estimations, added expert rule packages and made the mathematical modeling tool workable, accurate and convenient.
Centrifugal, helical, reciprocating compressor units and pumping stations –Clover PMM "knows" their peculiarities.
Clover works with "raw" data, complements and sorts it. And machine learning effectively finds relationships that are not obvious to humans.
Advantages of Clover PMM
Hybrid mathematical models
Describe sets of interrelated equipment
Identifies relationships in a large data stream that are non-obvious to humans
Predicts breakdowns within a space of 2 weeks
Integration into IT systems
Seamlessly integrates with any enterprise management software
Diagnostics and forecasting results presented in graphs, tables and visual reports
Implementation in industrial enterprises takes 2-3 months, and the effect of the system is observed after 3-4 months
TOP-20 "weak" places in oil and gas equipment
Based on the data collected, Clover experts determined points that most often fail during operations at oil and gas enterprises. Using Clover PMM, you can learn in advance about equipment breakdowns and sudden stops.
Helical compressor unit
Electric submersible pump unit
Centrifugal compressor unit
Sucker-rod pumping unit
Oil producing company
The company assigned Clover the task of increasing the energy efficiency of pumps at the enterprise. Pumping stations operating in the reservoir pressure maintenance system are an important part of oil production. The depreciation of the plants has an impact on the energy efficiency of the entire production. One of the tasks was to calculate the optimal operation mode of the cluster pumping station.
Clover PMM experts took about two months working on pre-project research and collecting data at the enterprise. The study was complicated by the partly-unreliable operation of the sensors. The design data also included varying pumping modes.
The Clover PMM solution allowed to pick out the correct data values. During the implementation of the project, a solution was found for calculating the optimal layout of pumping stations taking into account actual depreciation and electricity consumption. In such a case, Clover PMM takes into account the mutual effect of pump operations at the station.
daily savings on electricity
Clover PMM predicts the load on the pumps, based on the technical condition of the plants, and gives recommendations on the optimal loading of pumping units and the choice of their operation mode. The system monitors the efficiency-reduction trend and predicts the appropriate time for pump overhauls, which allows to plan maintenance.
27 mathematicians and engineers
develop physical and mathematical models for each type of equipment, study the boundary processes
create software that integrates into the IT-infrastructure of the production company
13 industry experts
study the equipment and submit their expert advice to mathematicians, business analysts in the building stages of the MX-models
53 scientific institutions
advise on sophisticated equipment of the oil industry, including the machine learning process