[Hydraulic Circuit Simulation Free
Facunda Ganesh
This software comes FREE with our hydraulic training courses . It includes a range of pre-built circuits that users operate to fully understand how they work. Change the loads or component settings to understand the effects on performance. Diagnose built-in faults .
The software is available for FREE to multi-user license holders via the teachers resource page. Individual download licenses are NOT included in the website subscription but can be purchased through the software links. Activation codes will be sent within 2 days.
Our circuit builder simulation package and virtual hydraulic test rigs are not the same as AmeSim, FluidSim, or similar simulation packages. They cost and perform quite differently. Users must select the one that best fits their needs and almost certainly start with our program before moving on to the dynamic simulation packages as they progress. It's likely that switch will only happen at advanced maintenance technician to experienced design engineer stage.
The key difference is that the advanced simulation packages are fully dynamic which require complex models for valve switching and fluid compliance. Our program only uses steady-state operation with basic valve sizes and setting variables.
With dynamic simulations it's far too easy to use incorrect parameters, such as valve leakage, which leads to incorrect results. These programs are best used by experienced design engineers who understand the complex component detail and have the time to build competent circuits that will avoid potential performance errors. Students new to hydraulics are likely to spend more time learning the program than the hydraulics and very few people use these programs in industry.
Our circuit builder is basic enough to work as a website app with no extra cost or time to install. Students with minimal knowledge can open and test example circuits or edit component values using a simple educational game interface, and little chance of entering 'silly' values. This product is designed as an easy to use teaching aid for people up to maintenance technician, or system design level.
If the model is properly constructed you can start a simulation. Hydraulic circuits can be simulated with all the available integration methods. In practice the default integration method (BDF) will give the most accurate and quickest response. Take care with hydraulic circuits that will give high pressure peaks and small flow rates. These circuits are sometimes hard to simulate. In practice high pressure peaks should be reduced with pressure relief valves to avoid damage. Fortunately adding pressure relief valves will in most cases improve the simulation response.
Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.
Engineering a successful stimulation system in the subsurface requires understanding rock response to different pressure conditions. The challenge is to predict as accurately as possible the growth and propagation of fractures in rocks during high-pressure fluid injection, a process called hydraulic stimulation. Several 2D and 3D tools developed by universities and industry are available to numerically simulate growth and predict fracture patterns for any geological condition. Due to the lack of analytical solutions for these coupled, non-linear processes, the accuracy of these simulation tools in solving hydro-mechanical processes can only be assessed by comparing the simulated results against laboratory-scale or field-scale datasets. A code comparison study investigating the capabilities of numerical codes and the quality of the numerical solutions to specific EGS problems is reported in1, where real field data was used to benchmark a number of simulators under specific assumptions. Indeed, achieving a well-constrained field-scale hydraulic fracturing dataset is almost impossible due to the unknown geological complexity at depth, which requires major assumptions regarding boundary conditions and subsurface parameters. Therefore, conceivably well-controlled and repeatable laboratory-scale experiments are indispensable for verifying model assumptions and constitutive relationships used by numerical codes to support EGS design.
To obtain meaningful results from laboratory tests, it is necessary to perform experiments on sufficiently large samples so that a stable fracture propagation can be achieved. Experiments of this scale are rather expensive and relatively rare in the geothermal energy sector, especially when performed under true triaxial conditions, i.e. with confining stresses along all the three axes. A non-exhaustive list of true triaxial facilities available worldwide, which can accommodate large samples, is presented in Table 1. However, most of these facilities are often commissioned by oil and gas companies, and therefore their results are not publicly available to the broad scientific community.
In this paper we present hydraulic stimulation datasets from experiments where the borehole axis was parallel to the minimum horizontal stress direction and therefore the plane of crack propagation was parallel to the maximum horizontal stress direction. Investigated rock samples represent the reservoir rock types of a potential EGS site in Mexico. The datasets represent hydraulic stimulation responses in quasi-homogenous and extremely heterogeneous crystalline rock types, such as, a very fine-grained granite and a coarse-grained marble, respectively. Additionally, a dense network of 32 acoustic sensors, comprising of 28 GmuG standard ultrasonic sensors ( ) and 4 Glaser amplitude-calibrated sensors4,5, was used to track the fracture propagation in real time.
The validation dataset includes fluid injection rate, pressure in the injection interval, confining stresses, mechanical and petrophysical properties of the rock specimens, properties of the injection fluid, mechanical details of the experimental set-up, and acoustic emission data. Additionally, we provide a Python-based code which can be used for processing the seismic data and visualizing of the experimental results.
Schematic figure of the triaxial set-up: components of the loading system (left); details of the slots for the acoustic sensors in the loading plates and a cross-section through the sample showing the borehole and the packer configuration (right).
The six surfaces of the samples were alphabetically numbered from A to F. A Cartesian coordinate system was used to systematically describe the samples. The origin of the coordinate system is located in the center of the surface A, the top of the sample. The z-direction corresponds to the borehole axis and points from the upper side A towards the lower side F. The y-axis points from surface E to C and the x-axis is positive from surface B to D (Fig. 1).
The liquid tank valve A and the bleeding line valve C (Fig. 3) were closed during the experiment and the volume of the pump cylinder is reduced to pressurize the injection system. The pressure was measured using two pressure transmitters p1 and p2 located at the upstream and downstream sections of the injection interval. In order to account for fluid viscosity and improve the measurement accuracy, the pressure of the injection interval was assumed to be the average of p1 and p2.
The injection fluid used in the experiment is a mixture of 98% glycerol and 2% ink. The red ink was added to the injection fluid to mark the propagation of the fracture and measure the fracture radius after the sample is split along the fracture plane. The dynamic viscosity of the injection fluid is highly temperature-dependent and therefore characterized in a separate test for the temperatures of the experiments using a rotary rheometer. The viscosity values of the injection fluid were then derived for each experiment from the temperature recorded during the experiment. The air content in the injection system was estimated based on the method described in6.
The control unit consists of a dedicated PC equipped with four M2i3122 12-bit 8-channel simultaneous sampling analog-to-digital (ADC) cards and one M2i6011 14-bit digital-to-analog (DAC) card synchronized via M2i Star-Hub technology. The data acquisition was controlled via GMuG software, which allows the configuration of each sensor channel setting and stores the acquired data in binary format both in the standard SEG-Y and in a proprietary GMuG format.
Once the first arrivals are picked for every transmitter-receiver pair, their distance was plotted against the P-wave arrival time. A linear regression of all the events provided the average P-wave velocity in the sample which is the value that minimises the sum of squared errors. Signals corresponding to transmitter-receiver pairs which were on the same side of the sample were discarded in the linear regression to avoid false picks related to surface waves.
Details of the experimental protocol steps are summarised below, whereas the salient stages are graphically depicted in Fig. 4. Example of a pressure response curve versus injection rate during different stages of an experiment is shown in Fig. 5.
All the valves A, B and C are opened and the injection fluid is circulated through the system (from the glycerol container to the bleeding line) to de-air the pipelines and saturate the injection interval (Fig. 3).
A leakage test is carried out to assess the tightness of the injection system. Valves A and C are closed and the injection system is pressurized to 3 MPa for approximately 600 s (Fig. 5). The required fluid volume which has to be injected into the borehole to maintain constant pressure is determined and used as an indicator of the tightness of the system.
795a8134c1